v

,1. '
(“$113, -.1 .'

.‘1';
' :1, .
11,1 “551“?“ ' :1“
:1‘$."L"":‘ ::
"Ih'5‘c' '.. “0151:" 0‘

 
    

          
   
  
 
 
 
  

.1" ,"'\"‘- , 1,». I.'u"’:"’
1.1.1.111?“ m,“
‘w "“1le 5‘ ‘1".\

'f‘." ”‘5 .« :"1,

 

g ($53511: 1,1511!

,1""‘u'5" 1.15 . , .
" 511515131 1.; 1 111 '

wJ'I'I- '5"

'u v 4. .3 I. .-

HJ 41113‘?\\1‘1I1:1]""-‘11' '1'“ {'3 (”neg-1'4

‘5," 1 h ; 115‘ .

_|..; _b\" 15.14,. \:‘\ 3131'“? '. . 4/11.}
' - U‘I "\\|UI '

  
  
  

 

__.‘.". .;. '
‘ .

-‘.‘ 'V
' ‘.. p.
-93"-

.‘7

111.4511“? @111
.1 ' "‘ 5:1

       

 

  

1,.“ '
Md 1
, '.‘_ '5‘55'111'51'1"“,1 '11,; \1'“
1 12"."11' 1.11 ""1‘ 54;. “MM
"U»:"1'.'1:1'"“/o "

25.1.1.5 ("1: -- .1 ‘42'1'1' ['3‘ 11:11., _'5.
'11“ “"""515J" .‘J,1

4511511‘1"')E1‘:'- 1,,I' -‘
1111511 m1 1.
a 1‘1 21 5:5

' ARM!» '55:. 1 ”M')?‘
‘5 "" "i" 541 3".1155'

‘..?fil :‘E; (:HM . J0I|1":1I

1,; 11,1."

   

 

         
 
 

  
       
   

    
     
     
   

 

1 .1
fl
, 1'”! 1.

      
   

   
     

'. J. ‘ .‘ . .11. ’ ‘ f" '1' "- l

- w: .. 1111-111- ‘ ~ ~. ~- - .
.,,-.W:..1-115 '. . '” ‘, 3‘ ”'7?“ 1-9.1- : 1 " ' " ' . P
11-11%,‘3111'1 ""1" in" 'I‘-“‘5 {1.11 11555? '.-" ,i' 1" 1‘" 1 1-51-54”? A
u. [ 3‘3' 1“ I131} :1%:' ' 1 . ' .. '11:"qu 1.... ,. . ..

1 U ‘(1, 1"": ‘51-‘11 "0.01515, .,5 "U" 571.11 151?}.11'1 (k . I . L , ' , H ,1 } ,,.1",-i'1':."'-' 5,531! 1-1;.114'11‘5 1.11; 5 1“,"." .. {"3
5555 1'11 (1“... _ 11'“ ‘ ‘1 .. ' ' ' "- . . I: , .. 'V" 1:19:515131'fiv'511"'5M"':I"" I. ‘

‘I
If"!
1

   

     
 

 
 
  

    
  
 
 

Ln (‘1‘:

1.1595 ‘

  

' "‘11 ""u5'. W” 511"“

1.1. "" 1'15..'..;:'i 11-5-5"

4 V'l "1'“:|".""
l {"11‘ u
I“? ‘31P, 11:15 """5

5|I,"'l
I

    
 

l
1""1U‘.:'."1'l"

5'5)”

"4
l

1%“
1 11111 1.1.1.1. . ‘
I 1 1 1111113555. 115 ' I 'I' 1" 1"":1, 0
J r,./,, ..;'. 1,1 , 1 ., 5" ,... .
5. 1 1.15 “MM-1‘1, 11:1; 1:11.11: “‘ '* ‘51"... '1=

 

 

flESlS M

a IBRARY
mailman Sta te
‘ University I

This is to certify that the
thesis entitled

SENSORIMOTOR INTELLIGENCE, LANGUAGE
DEVELOPMENT, AND PARENT-CHILD INTERACTIONS
IN l4, l8 AND 22 MONTH OLDS

presented by
CHARLES TALOR

has been accepted towards fulfillment
of the requirements for

M.A. Psychology

degree in

 

 

fl/A

Major professor

ThomasH Carr
Date 12/21/82

 

0-7639 MSU is an Ajfirmative Action/Equal Opportunity Institution

 

 

 

 

 

 

'PV1ESIHJ BEIDRNING MATERIALS:
Place in book drop to
LJBRARJES remove this checkout from
1—— your record. FINES_ will

 

 

be charged if book is
returned after the date
stamped below.

W “‘..-‘1

 

”w”-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

SENSORIMOTOR INTELLIGENCE, LANGUAGE DEVELOPMENT,
AND PARENT-CHILD INTERACTIONS IN 14, 18 AND 22
MONTH OLDS
by

Charles Talor

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF ARTS

Department of Psychology

</

/

2)

C

sass-1

ABSTRACT
SENSORIMOTOR INTELLIGENCE, LANGUAGE DEVELOPMENT,
AND PARENT-CHILD INTERACTIONS IN 14, 18 AND 22
MONTH OLDS
by
Charles Talor

Thirty male and female infants 1h, 18 and 22 months of
age were assessed for sensorimotor intelligence, language
development, and mother speech interactions in videotaped
situations. Object permanence development related only
to certain object word vocabulary categories in comprehen-
sion that reflect knowledge of objects. Means-end relations
development related to verb and marginally related to caus-
ality statements in infant speech. in a manner consistent
with Bates (1976) observations. Situations affected mother-
infant speech relations and infant utterances, though not
infant MLU. Many forms of mother speech related to child
speech. The general pattern is that inherently complex
mother speech negatively related to child speech; fine
tuned speech positively related to child speech. Some mother

speech related to child relational comprehension.

ACKNOWLEDGEMENTS

I would like to thank the individuals who richly deserve
my thanks and appreciation for their help on this project.

The data collection, coding and mother interviews were
immensely aided by the Psychology 490 students who worked
with me. Also, the mothers of the children in this study
were critically helpful in data collection--especially in
keeping diary records of their children. Without their
help, this project could not have been completed.

I would like to thank the members of my thesis committee
for their contributions to this project. Tom Carr provided a
model of competence and dedication to the research and advisory
process that I'm sure will stay with me throughout my career.
Helen Benedict introduced me to the interesting fields of
language development and cognitive growth. Ellen Strommen
provided valued support and advice for my work.

Finally, my fiance Suzanne Ewing and our friends were of

tremendous value for their support and encouragement.

ii

TABLE OF CONTENTS

LIST OF TABLES . . . . .

Chapter
I.

II.

III.

IV.

Introduction . . . .

Nature of the Problem . . .

Review of the Literature . .
Language Development . . .
Cognitive Development . .

Cognitive and Language Development
Environmental Effects on Cognition
and Language . .
Summary . .

Statement of Purpose

Hypotheses .
Pilot Study . . . .
Introduction . . . .
Methods . . . . .
Subjects . . . .
Procedures . . . .
Measures . . . .
Subject recruitment and
Confidentiality . . .
Results . . . . .
Discussion . . . .
Study II . . . . .
Introduction . . . .
Methods . . . . .
Subjects . . .
Subject recruitment and
Confidentiality . . .
Procedures .
General Procedures for Sessions
Measures . . . .
Videotaping procedures . .

Data Reduction and Coding .

iii

page S

V. Results . . . . . . 57
Introduction . . . . . 57

Transcript Coding and Consistency . 57

Means-end Relations Scale Consistency 58

Hypothesis One . . . . 61

Hypothesis Two . . . . 63

Hypothesis Three . . . . 65

Hypothesis Four . . . . 70

VI. Discussion . . . . . 78
Hypothesis One . . . . 78

Hypothesis Two . . . . 80

Hypothesis Three . . . . 85

Hypothesis Four . . . . 89

General Summary and Conclusions . 93

REFERENCE NOTE . . . . . . 94
LIST OF REFERENCES . . . . . . 95
APPENDICES . . . . . . . 99

Appendix A. Language Comprehension
Assessment . . . 99

Appendix B. Object Permanence Assessment 102

Appendix C. Table A . . . 106

iv

Table
Table

Table

Table

Table

Table

Table

Table

Table

Table

Table

10

11

LIST OF TABLES

Object Permanence Performance
Means-end Relations Performance

Toy Preferences during a five minute Free-
Play

Partial Correlations of Object Permanence
Performance to Number of Words in Children's
Vocabulary Categories

Partial Correlations of Object Permanence to
Children's Speech in Free-Play and Videotaped
Situations

Partial Correlations of Means-end Relations
Performance to Children's Vocabulary
Categories

Partial Correlations of Means-end Relations
to Children's Speech in Free—Play and
Videotaped Situations

Partial Correlations of Mother Speech to
Child MLU and Utterances in Videotaped
Situations

Correlations of Mother Speech to Children‘s
Verb (V) and Noun (N) use in Videotaped
Situations

Mother Speech Relations to Children's
Auxiliary (A) and Causality (C) use
in Video Situations

Multiple Rs (MR) and Beta Weights Predicting
Children's Speech in Videotaped Situations
with Age, Object Permanence, Means-end Rela-
tions, Mother Yes/No Question and Imperative
Use as Predictors

38
39

#1

6A

65

68

69

71

72

73

76

INTRODUCTION

Recently research in language development has focused
on the first appearance of language comprehension and pro-
duction (Benedict, 1979; Nelson, 1973). Language development
has also been compared to cognitive development (Rodgon,
1976), and social interactions (Bruner, 1978).
Nature of the Problem

Recent studies of initial language and concurrent cog-
nitive development emphasize one form of sensorimotor (men-
tal) structure (object permanence) and compare that structure
to one aspect of language development (verbal utterances).
These studies find increases in language complexity are re-
lated to the development of representative thought as marked
by successful "invisible" search for objects (object perman-
ence stage 6). Also, the typical study of social interac-
tion and language relates grammar and syntax development in
young children to mother speech and has not emphasized simul-
taneous relations of cognitive development and mother speech
to child language development. Overall, the pattern of rela-
tions between language development, cognitive development,
and social interaction has been extensively mapped out with
measures of object permanence and verbal utterances, while
receptive language, other forms of sensorimotor growth, and

1

2

mother speech have not yet been interrelated to the growing
body of work examining object permanence and verbal language.

This paper focuses upon} (a) distinguishing receptive
language abilities from productive language abilities,
(b) examining sensorimotor growth in areas other than object
permanence development, (c) reviewing findings which relate
sensorimotor growth to language development, and (d) relat-

ing maternal language factors to infant language.

REVIEW OF THE LITERATURE

Language development

Language development research has emphasized language
production (Bloom, 1973; Bowerman, 1973; Brown, 1973;
Greenfield & Smith, 1976; Nelson, 1973). though recent
findings indicate differences between receptive and produc-
tive language abilities (Benedict, 1979; Huttenlocher, 197A).
Receptive language abilities begin on the average at 10
months of age (Benedict, 1979). 2 months before productive
language begins (Nelson, 1973). Also, initial acquisition
rates of words in comprehension are greater than production
(22.03 versus 9.09 words per month, Benedict, 1979). This
means receptive vocabularies are generally larger than pro-
ductive vocabularies during the early stages of language
development (Benedict, 1976).

Early productive and receptive language also differ in
the way word meanings are related. The understanding that
words can alter the meaning of other words develops late,
occurring 6 to 10 months after receptive language begins.
In contrast, words that may alter the meaning of other words
have been described as appearing early in speech production
(Greenfield & Smith, 1976), though the appearance of such

words in production takes place after the comprehension of

these same words.

Words that can change the meaning of other words have
been termed "relational" (Benedict, 1976). Relational words
are often, though not always, modifiers. Modifiers make up
only one type of word category that characterize early vocab-
ularies. Frequency counts of words belonging to different
word categories can distinguish receptive and productive lan-
guage abilities. Receptive vocabularies tend to have a
greater proportion of action-oriented words than productive
vocabularies during the acquisition of the first 50 words
(36% versus 19%. P <.01, Benedict, 1979).

The larger proportion of action-oriented words in recep-
tive vocabularies may be due to a general action-oriented
response strategy by the children (Shatz, 1978). Shatz
(1978) found that when 19- to 34-month-olds gave meaningful
responses to linguistic commands, the responses were predom-
inantly action-oriented although many commands did not re-
quire overt physical responses. The frequency of action—
oriented responses dropped as older infants were assessed,
though action-oriented responses predominated at all ages.
For example, when children were asked to identify the loca-
tion of a toy, most children who responded overtly pointed or
picked up the toy in question rather than verbally identi-
fying the toy's location.

Comprehension and production vocabularies also differ in
other word category frequencies. Comparisons of the first

50 words in comprehension and production vocabularies show

5

fewer general nominals in comprehension compared to produc+
tion (39% versus 50%, p < .01), fewer personal-social words
(5% versus 10%, p < .01), and more specific nominals (17%
versus 11%, p < .01).

Overall, substantial differences between comprehension
and production of early language have been found in studies
of vocabulary acquisition (Benedict, 1979: Nelson, 1973), and
the comprehension of multiword utterances by young children
may be relational or non-relational. Measures of both recep—
tive language and productive language can enhance our picture
of early language competence compared to using each measure
separately.

Qggnitive Develppment

Piaget (1952, 1954, 1967) describes early cognitive de-
velopment in terms of action schemes and cognitive influences
on action schemes. Action schemes are the actual set of be-
haviors possessed by a young child to perceive and manipulate
objects, such as eye scanning patterns in looking. Object
manipulation characterizes for Piaget a child's understanding
of objects and their properties. The proposal that cognition
influences action schemes emerges from Piaget's philosophical
position that the child actively seeks knowledge about the
environment. These cognitive influences can be considered as
procedures or strategies which determine action scheme usage
in particular situations.

Early in development a child tends to use a limited set
of behavioral acts on objects. A child acquires a larger set

of acts with age and experience. The early use of action

6

schemes is often haphazard. The child must apply several
different behaviors until an appropriate one is found for
a particular situation. Development occurs in the adjustment
and adaptation of old behaviors to meet the demands of new
situations. The result is an expanding and increasingly
differentiated set of situation-appropriate behaviors. Ad—
justments begin initially after old behaviors fail in new
situations. However, long term development is more than the
acquisition of larger and larger sets of situationally-
appropriate behaviors applied in a trial-and-error fashion.

In addition to acquiring larger sets of behaviors, a
child eventually "reasons out" appropriate sets of actions
without carrying out any "preliminary" acts. A "planfulness"
develops as correct behaviors are devised systematically and
directly without feedback from the failures of unadjusted
behaviors. This planfulness is a major development for
Piaget, indicating the first use of internal, mental repre-
sentation by the child. For example, Bates (1976) notes a
young child carrying flowers and wishing to enter her house.
The child places her flowers on the porch and began to open
the door when she looked down at the flowers and stopped.
She then moved the flowers outside the path of the door, cor—
rectly anticipating the effect of its movement on her deli—
cate flowers.

Piaget calls the period in early cognitive development
during which a child first acquires a set of behaviors to act

intentionally on objects as sensorimotor stage 4. When a

7

child adjusts established behaviors to meet changes in new
situations, the child is said to be operating in sensorimotor
stage 5. When the child "plans" appropriate behaviors in
situations before acting the child is seen as operating in
sensorimotor stage 6. Thus the last three stages of the sen-
sorimotor period, beginning at about 8 months of age and car-
rying on through age 18 to 24 months on the average, mark the
development of strategically planned, goal-directed action.

Piaget describes a hierarchical sequence of stages.
However, he notes that complex situations which exceed a
child's capacity for understanding tend to cause that child
to revert to less advanced strategies to cope with the situ-
ation. Similarly, a stage 6 child may be unable to "plan"
an appropriate course of action without attempting some
"trial" behaviors in difficult situations.

The apparent interaction between situational complexity
and the child's level of cognitive reasoning is explained by
Piaget as horizontal decalage. In a horizontal decalage.
children operate at an advanced stage of cognition in simple
situations, but at a lower stage in complex situations.
Piaget organizes the decalages around four major situational
interactions. These interactions involve cognitions associ-
ated with objects, space, causality, and time. The cogni-
tions for each interaction are thought to be interdependent.
Knowledge of objects is necessary for the understanding of
space and causality. Causality knowledge is critical for

the understanding of time.

8

Uzgiris (1973) confirms Piaget's general findings of
stages and horizontal decalages. Uzgiris (1973) found that
new knowledge of causality and space developed after critical
developments in object knowledge. This is understandable as
space understandings require knowledge of objects in space,
and causality requires a knowledge of how actions and objects
can create movements and activities in other objects and
people. However, Uzgiris (1973) found that stage 4 and stage
5 behaviors were difficult to distinguish from one another,
and two distinct stages were found in the behaviors associ-
ated with stage 6 performance.

The implications of Uzgiris's findings are first, early
behavior adjustments of stage 5 may be difficult to distin-
guish from non-adjusted behaviors of stage 4. Infants may
simply try a variety of "trial and error" behaviors to solve
a situational problem, rather than try one behavior which
produces unsatisfactory results. Second, the "reasoning"
that a child performs may be divided into "simple reasonings"
of an early stage 6, and "harder reasonings" of a late stage
6. Uzgiris (1973) reports space and causality behaviors
were not complex when children solved invisible hidings
(early stage 6) in a test of object knowledge (object perm-
anence). Space and causality behaviors increased in complex-
ity only after the children solved multiple invisible dis-
placements (late stage 6) in the same test of object know-
ledge. For example, space behaviors of young children in

early stage 6 include retrieving out of reach objects with

9
small sticks, while in late stage 6, children recognize and
do not stack onto a stick a ring with its hold filled which
is mixed among several unfilled rings. Late stage 6 behav—
iors are more subtle than early stage 6 behaviors.

Piaget's situational interactions and Uzgiris's findings
are important for two reasons. First, many studies of sen-
sorimotor development measure only one sensorimotor activity
(usually object knowledge, i.e. Corrigan, 1978). Second,
sensorimotor development may be distinguished into early and
late stage 6 functioning. Since different sensorimotor ac-
tivities mature at different times and rates under different
conditions (Uzgiris. 1973), individual differences are im-
plied in cognitive development. Individual differences in
cognitive development may relate to individual differences
in language development if a cognitive primacy view of lan-
guage development such as that of Sinclair (1973), covered
in the next section, is true.

Cognitive and Language Development

Recent research has applied the theory of sensorimotor
development (Piaget, 1954) to explain early language devel—
opment. Ingram (1978) and Sinclair (1973) believe that early
language behaviors are manifestations of sensorimotor action
schemes and reasoning abilities. Empirical support for the
sensorimotor view of language development, however, has been
inconsistent.

Rodgon (1977) examined relations between six measures
of cognition and one measure of language in a longitudinal

assessment of eight 16- to 21-month—olds. The cognition

10

measures were symbolic play, deferred imitation, role
playing, object permanence, cup manipulation, and object man-
ipulation. The language measure consisted of word production
vocabularies. Infants were matched on verbal ability by
being able to speak one, but not more than one word at a
time. The cognitive measures were given once during the
first few months of study while language productions were
assessed continuously.

Rodgon found no overall pattern among the cognitive
measures and the ability to speak one word utterances. How-
ever, all children showed symbolic functioning in at least
one cognitive measure, suggesting several explanations.
First, cognition and symbolism in language depend on one
another for development. This effect may be uni-directional.
Concepts assessed in cognition may aid acquisition of sym-
bolism in language. Alternatively, language symbolism may
aid symbolic performance in cognition measures. Two-
directional effects may also occur. For example, object per-
manence knowledge may influence symbolism in acquiring object
words, while symbolism in social words may influence the
acquisition of symbolism in Rodgon's cognitive measure of
role playing.

A second form of relationship between cognition and lan-
guage considers the two as depending on a common underlying
capacity for abstraction. For example, symbolism involving
objects may facilitate object understanding as marked by

object permanence and object labeling. A third form of

11

relationship between cognition and language is independence,
in which symbolism in both areas develop separately but on a
similar course.

Unfortunately, Rodgon's findings on cognitive and lan-
guage relations are restricted by the exploratory nature of
her study. The study was longitudinal for language develop-
ment, while each cognitive measure was given once. Rela-
tions may have emerged in longitudinal assessments of both
language and cognition, or within a cross-sectional design
of several different ages. For example, words coding move-
ments of visible objects, such as "move," may occur only
after the child shows an understanding of multiple invisible
displacements in an object permanence task. Alternatively,
a child may show knowledge of objects not in the immediate
environment by speaking words such as "all gone" prior to
success in locating objects hidden invisibly in an object
permanence task. In a longitudinal study, these relations
will be readily observed, if present. In a cross-sectional
study, a pattern should emerge in which one behavior co-
occurs only as a subset of the second behavior across sub-
jects.

In the preceding examples of the acquisition of "move"
and “allgone,” the consistent observation of one behavior
before the other indicates a relationship between language
and cognition and also the direction of that relationship.
Though a consistent relationship would support the first

argument, the second argument can still be valid if both

12

behaviors are mediated by a common process but that one be-
havior is easier to produce and is observed earlier. Third
factors, such as the linguistic environment of children may
provide a new way to examine cognitive and language rela-
tions. Linguistic environment factors may alter the time
span between achievement of behaviors in one area compared
to the second area. For example, stage 5 object permanence
may precede the appearance of object words in a child not
living in a linguistic environment emphasizing object word
acquisition. Both behaviors may be reversed or appear simul-
taneously in a linguistic environment stressing object word
acquisition. Linguistic environment factors on child lan-
guage development will be described in a later section.

Rodgon's (1977) exploratory study also attempted to
equate language fluency by assessing children producing only
one word utterances. However, children speaking one word
utterances may understand multi-word utterances (Benedict,
1976), and the number of words in the children's receptive
vocabularies were probably greater than the children's con-
current productive vocabularies (Benedict, 1979; Huttenlocher,
1974). These differences in receptive and productive lan-
guage capabilities imply that assessments of the number of
units of linguistic information that can be operated upon at
one time, as well as the aggregate amount of linguistic in—
formation available to a child, is greater in receptive lan-
guage skills and that these receptive language skills should

be used with productive language skills to determine language

co
9“!
of
be

+..

 

 

13

competence.

The greater sensitivity of receptive language skills has
two related empirical implications. First, critical stages
of cognitive growth may be more primitive than previously
believed, such as stage 4 or 5 in object permanence. Second,
relationships between cognition and language should be
stronger for comprehension rather than production.

Rodgon's finding of no overall pattern of relations be-
tween cognitive measures and language may be related to the
use of exploratory cognition measures, especially symbolic
play and deferred imitation which received few or no re-
sponses. Future examinations of language and cognition may
be clarified by Rodgon's multi-measure approach, but based
on a multi-faceted theory such as the four sensorimotor sit-
uational interactions using standardized measures (Uzgiris
and Hunt, 1975)-

Zachery (1978) examined decalage among different aspects
of Piaget's sensorimotor stage 7 and language productions.
Twenty-four infants between 12 and 24 months of age were ad-
ministered five of the Uzgiris and Hunt (1975) scales: ob-
ject permanence, space, imitation, causality, and means-end
relations. Abilities measured by these scales occurred in
that order of difficulty with a Guttman scale coefficient of
.94.

Zachery's infants were compared on four general sentence

categories based on semantic units produced per sentence.

The four categories were: (a) two word utterances such as

14
naming or recurrence of objects, (b) utterances coding se-
mantic relations such as agent—action and nominal-locative,
(c) three term semantic relations such as agent—action-ob-
ject and agent-action-locative, (d) four term semantic rela-
tions such as agent-action-dative-object and agent-action-
object-locative. These 4 categories formed a perfect or-
dinal relationship of difficulty in the order presented.

Unfortunately, Zachery did not control for age factors
when comparing sensorimotor and sentence production abil—
ities. Sensorimotor abilities were reported as correlat-
ing .75 with age. In view of these limitations, Zachery's
findings will be interpreted cautiously.

Zachery found 22 infants achieved stage 6 performance
in object permanence; 16 infants produced one word utter—
ances though not all were stage 6 performers. Ten infants
produced two word utterances or sentences; these infants all
achieved stage 6 performance on at least 3 different scales.
Zachery's findings show a correspondence between sensorimotor
activity and language development, a correspondence which is
stronger at higher levels of sensorimotor activity (as in-
dexed by the number of scales passed at a stage 6 level) and
language production (such as sentences). High stage 6 per-
formance did not mean advanced language skills. These find-
ings support a hypothesis incorporating several aspects of
sensorimotor intelligence to the development of advanced
language.

Zachery's findings are in support of, but do not require,

15
but do not require a multi-faceted view of sensorimotor de-
velopment (Piaget, 1954) as the basis for language acquisi-
tion (Ingram, 1978; Sinclair, 1973). Multi-faceted views of
sensorimotor development have been developed by Bates (1976)
and Snyder (1978) who describe Piaget's object knowledge as
facilitating language intents of gaining attention from
others. A second form of language intent, which maybe used
with the first form, is language activity based on space
knowledge. This second form is directed toward getting ob-
jects and actions by using people as agents to get desired
objects or produce desired acts.

Bates (1976) conducted a longitudinally study of three
children from 3 to 24 months of age. Bates observed that
sensorimotor stage 5 goal oriented behaviors first appeared
at the same time as communicative gestures of attention
seeking and request making (for specific objects or actions).
When sensorimotor stage 6 goal oriented behaviors first ap-
peared, these communicative gestures were observed as being
replaced by verbal utterances of the same intent. For exam-
ple, a child repeatedly pointed to a kitchen sink to request
water from her mother, a gesture which appeared when the
child first used objects to extend reach. This use of ob-
jects for extension corresponds to a stage 5 understanding
of "secondary supports" (Piaget, 1954).

Bates (1976) findings are intriguing but must be quali-
fied, as standardized measures of sensorimotor activity were

not used. Snyder (1978) addressed this problem by using all

16

of the Uzgiris and Hunt scales with measures of gestural and
verbal language in controlled naming and request situations.
Snyder assessed 15—month-old language-normal and 36—month-old
language-delayed children; both groups had, on the average,
the same number of words in their production vocabularies,
and their MLUs were approximately one. Language-normal sub-
jects produced significantly more linguistic imperative and
linguistic declaratives (Mg = 5.0 and 3.4) compared to the
language-delayed subjects(M§ = 1.4 and 1.06) across ten
trials. The language-delayed subjects produced significantly
more nonverbal and gestural "proto" imperatives and "proto"
declaratives (Mg = 7.13 and 4.8) compared to language-normal
subjects (Ms = 4.86 and 3.0) in the same ten trials. Gener-
ally, the language-delayed group expressed themselves non-
verbally, and the language—normal group expressed the same
intents verbally. Imperative and declarative use was not
correlated with object permanence ( Rs = .00 and .09). A
Stepwise multiple regression using all Uzgiris and Hunt
scale scores showed the means-end relation scale best pre-
dicted and differentiated the two groups on verbal and ges-
tural use.

Snyder's findings indicate verbal responding relates to
means-end relations development. All children in the study
achieved stage 6 performance in object permanence, but only
the language-normal children achieved stage 6 performance in
means-end relations. The means-end relations scale specifi-

cally measures secondary support use within the general area

17
of space knowledge (Piaget, 1954). The relationship of
means-end relations to verbal request making is consistent
with observations made by Bates (1976).

Developmentally, Uzgiris (1973) and Zachery (1978) re-
port findings similar to Snyder (1978), in the sense that
object permanence develops before means-end relations. How-
ever, Uzgiris's (1973) distinction of early and late stage 6
in object permanence qualifies Snyder's findings since
Snyder did not put into her regression analysis a fully dif-
ferentiated object permanence measure, effectively restrict-
ing its range. Whether this restriction of range was suffi-
cient to reduce the relationship of object permanence to the
criterion variable in the multiple regression solution, and
hence its final order in that solution, is not known.

Whether early speech is related to means-end relations
or object permanence is not known since studies using differ-
entiated object permanence scales that show relations to
early speech (Corrigan, 1978: McCune-Nicholich, 1981; Ramsay,
Campos, & Fensionq 1978 ) did not also assess means-end
relations. These studies indicate that acquisition of cer—
tain speech forms and vocabulary are related to object perm-
anence development.

Ramsay et al examined longitudinally early and late
stage 6 object permanence and object word development in 24
infants from 10 to 21 months of age. Entry into early stage
6 preceded acquisition of the first object word by approxi-

mately one month (Y = 10.9 versus 12.4 months, p < .006).

18
Entry into late stage 6 preceded a rapid acquisition of ob—
ject words (Mg = 17.1 versus 17.9 months) for most subjects
(gs = 17 versus 7, p «<.05).

Ramsay et al's findings show strong links between ob-
ject permanence development and object word activity, though
the very few infants with accelerated object word acquisi-
tion rates before late stage 6 argue against an object per-
manence position. These atypical infants may be the result
of measurement inaccuracies or due to the emergence of a
second form of cognition, such as means-end relations.

Corrigan (1978) conducted a longitudinal study of pro-
ductive language and object permanence development in 3 chil-
dren from 10 to 22 months of age. The children increased
their "language scores" after entry into early stage 6. The
"language scores" were a combination of MLU and any other
verbal output produced by the children. Words representing
nonexistence (e.g. "allgone") were spoken after entry into
late stage 6 in all children. Two of the three children
spoke words representing recurrence (e.g. "more") after
achievement of late stage 6. One child used the word "more"
prior to late stage 6, but this use was limited to objects
present to the child (e.g. "more raisin", "more top” meaning
spin the top). This third child's use of the word "more"
changed after entry into late stage 6, when "more" was used
to request objects and actions not present to the child, a
use of "more" comparable to the other two children.

McCune-Nicolich (1981) examined in a longitudinal study

19
object permanence and productive language development in 5
girls from 14 to 24 months of age. Most of the girls (4 of
5) produced words coding nonexistence while they were in
early stage 6, and 2 of the 5 girls produced words coding
recurrence while in early stage 6.

The studies of object permanence and language develop-
ment agree that achievement of object permanence stage 6 is
related to increases in language production. Early stage 6
was linked to acquisition of object words, increased verbal
activity, and words coding nonexistence and recurrence.
While Corrigan reports acquisition of these latter two word
classes in late stage 6 in contrast to McCune-Nicholich's
report of acquisition during early stage 6, the difficult
and abstract nature of these two word classes and acquisi-
tion by some but not all of McCune-Nicholich's subjects
during early stage 6, indicate individual differences on the
production of these words. Both studies find these abstract
words produced in stage 6, and not an earlier stage.

Throughout this paper, language development has been
related to one or more aspects of sensorimotor development.
Since studies using differentiated object permanence measures
did not also measure other aspects of sensorimotor develop-
ment, such as means-end relations development, verbal respon-
siveness may be related to either sensorimotor activity.

For example, Corrigan's observation of increased verbal ac-
tivity with early stage 6 object permanence development con-

trasts with Bates's and Snyder's observations of increased

20
verbal responsiveness and request making with stage 6 means-
end relations development.

Though cognitive relations to language have been empha-
sized, another factor, environmental influences, may affect
language and cognition. Environmental influences may ex-
plain some differences found by Corrigan, McCune-Nicholich,
Ramsay et al, and Zachery. Corrigan had 1 of 3 subjects pro-
duce the word "more" prior to late stage 6, though use of
that word changes when the child entered late stage 6.
McCune-Nicholich found some children acquired words repre-
senting nonexistence and recurrence in early stage 6.

Ramsay et al found some children acquiring object words be-
fore entry into early stage 6, as well as other children ex—
periencing rapid word acquisition before entry into late
stage 6. Zachery found complex sentence producers achieved
stage 6 performance on at least 3 sensorimotor measures,
though not all who achieved stage 6 performance on at least

3 measures produced complex sentences. To speculate on these
differences, variations may exist in the children's language
environments. Ramsay et al's subjects may have had many ob-
jects to label in their environments or had parents who en-
couraged acquisition of object labels. Parent encouragement
may be planned or structured using daily sessions emphasizing
object labeling, or parent encouragement may be without aware-
ness, such as a sharp decrease in imperative use around their
children. Different language environments may provide dif-

ferent forms of language acquisition opportunities in terms

21

of frequency and clarity of speech that interact with the
child's cognitive sophistication and communicative intent.
Possible environmental effects will be described in the next
section.
Environmental effects on Cognition and Language

Gross enviromental characteristics, such as the number
of hours per week of caregiver-infant contact or the number
and type of toys available to young infants, have been re—
lated to early cognitive development (Bayley and Schaefer,
1964; Bradley and Caldwell, 1976, 1980; McCall, Appelbaum, &
Gogarty, 1973; Paraskevopoulos and Hunt, 1971). Since the
effect of environmental factors on early cognitive develop-
ment has been well documented, this section will emphasize
environmental influences on early language development.

Furrow, Nelson and Benedict (1979) compared infant and
maternal language characteristics when the infants were 18
and 27 months of age. At 18 months of age, the infants were
roughly equivalent in verbal language skill, belonging with-
in Brown's (1973) stage one categorization with a MLU of
approximately one. Several mother-speech behaviors when the
infants were 18 months of age were significantly related to
subsequent infant speech behaviors at 27 months of age.
Higher frequencies of yes/no questions, interjections and
noun—to-pronoun ratios in mother-speech to 18-month-old in-
fants positively related to 22-month-old infant speech char-
acteristics such as MLU, verbs per utterance, noun phrases

per utterance, and auxiliaries per verb phrase. On the

22

other hand, higher frequencies of words, verbs, and pronouns
by mothers to their 18-month-olds negatively related to 27-

month-old infant language, such as MLU, verbs per utterance,
and noun phrases per utterance.

The correlations found by Furrow et al between mother—
speech to infants at 18 months and later infant speech at 27
months did not explain the correlations between concurrent
mother and infant speech when the infants were 27 months old.
When the infants were 27 months old, mother's use of imperi-
tives and verbs positively related to child MLU, verbs per
utterance, and noun phrases per utterance, while mother in-
terjection use negatively related to the same child language
variables. The number of significant correlations between
mother-speech when the infants were 18 months old to infant
speech at 27 months (20 out of 60 reported), was greater
than the number of significant correlations of mother—speech
to 27-month-olds to 27—month-old speech (11 out of 70 report-
ed).

Furrow et al's results suggest two classes of mother—
infant speech relations. First, some mother-speech to in-
fants facilitates or hinders later infant speech acquisition.
Second, some mother-speech relates to current infant speech.
However, Furrow et al contend that most mother to infant
speech has long term effects. Though mother-speech is modi-
fied for current infant comprehension producing current sta-
tistical relationships, these modifications reduce speech

complexity, and reduced complexity is hypothesized as

23

facilitating long term language growth. Thus, mother frag-
ment and "deleted frontal auxiliary yes/no question" use at
18 months have simple sentence structures correlating posi-
tively to infant MLU at 27 months (rs = .57 and .77). These
simplified speech productions still contain complex language
elements. For example, verbs were defined as complex ele-
ments, since verbs have no specific referent. Furrow et al
found mother verb use at 18 months negatively correlated
with infant MLU at 27 months (r = -.71).

Newport, Gleitman, and Gleitman (1977) contend most
mother-to-infant speech is for current communicative needs
and has little long term effect. Newport et al examined
mother-speech to their children in three age groups (12 to
15 months, p = 3; 18 to 21 months, p = 6; 24 to 27 months,

p = 6), and related this speech to infant speech 6 months
later. Partial correlations, removing variability of infant
speech and age during the first assessment, showed few signi-
ficant relations. Mother use of yes/no questions positively
related to "auxiliaries per verb phrase" (r = .88, p < .001)
and marginally to MLU (r = .50, p <.1). Repetition nega-
tively related to child "auxiliaries per verb phrase" (r =
-.50, p < .1). Only 4 out of 50 relations were significant
(10 out of 50 at p <.1). Newport et a1 conclude mother-
speech to infants shows relations for acquisition of a few
language forms, such as infant auxiliary use which was re-
lated to mother "yes/no question" use which emphasizes fron-

tal auxiliaries. Newport et al suggest long term language

24

acquisition is due to increased language processing by in-
fants. Such processing may be due to cognitive development
or maturational factors.

To speculate on Furrow et al's and Newport et al's find—
ings, the pattern of relations may reflect the child's lan—
guage competence and mother language facilitation effects.
However, the 18 month old is largely non-verbal, yet all
child language measures were verbal. NewPort et al used
partial correlations to correct for age and language differ-
ences in their sample. Furrow et al used simple correlations
since their children were about equal in verbal language (MLUs
about one). This emphasis on verbal language may mask under—
lying differences in receptive language. Furrow et a1 note
that while their children were approximately equal in verbal
skill, the 18 month olds differed in receptive language
skills. These receptive language differences have been de-
scribed by Benedict (1976; 1979) and include vocabulary size
(ranging from approximately 50 to 150 words), and the number
of words understood in an utterance (ranging from approxi-
mately 2 to 4 words). Unfortunately, receptive vocabulary
size and the number of words comprehended at one time were
not compared to any of the mother or child speech variables
reported in the Furrow et al paper. Bohannon and Marquis
(1978) report adults will increase their MLU to a 32 month
old who appeared to understand their speech, and correspond-
ingly decrease their MLU when the 32 month old appeared not

to understand their speech. Receptive language differences

25

in Furrow et al's 18 month olds may have been related to
their 27 month old speech. Mother speech to 18 month olds
may then relate to 18 month old comprehension and not to
their speech at 27 months.

Relationships between infant language competence and
mother language behaviors may appear in several areas.
First, mother word categories may influence acquisition of
certain words and word categories in speech and comprehen-
sion, such as causality concepts in speech (Hood & Bloom,
1979) and auxiliary use in speech (Newport et al, 1977).
Second, mother may differ in particular objects, actions or
people they wish their children to know (Nelson, 1973).
Third, mothers may play interactive games with their children
which may differ in frequency and content (Lieven, 1978;
Ninio & Bruner, 1978; Ratner & Bruner, 1978). Interactive
games give infants exposure to conversational skills, such
as initiating speech or asking questions. These games are
also structured opportunities for the mothers to teach spe-
cific words and actions. Mothers may differ in the above
three areas which may relate to infant language skills.

Mother speech to their children does not occur in isola-
tion. Often there is a purpose or direction to the immediate
conversation that provides a structure or topic that the moth-
er or child uses as a reference point for speech.

The purpose of a situation can affect mother-child
speech in that situation. Children can be placed or asked

to engage in a "free play" with a variety of toys, with or

26

without mother participation (e.g. Benedict, 1976; Corrigan,
1978). Golinkoff and Ames (1979) compared mother—child
speech in "free play" and "structured" play situations, and
found mothers used more questions and less imperatives in
the free play situation and the opposite pattern in the
structured situation. Benedict and Truhon (note 1) compared
mother-infant speech in free play, book reading, and struc-
tured play situations, and found mother speech in structured
play compared to the other two situations as having: (a) the
greatest number of utterances, (b) the highest ratio of
imperative/ interrogative use, and (c) the lowest noun/pro-
noun ratio. Mother speech in structured play contained
speech associated with intents to change activity or non-ac-
tivity of the infant—listener, especially mother use of im-
peratives and questions.

Benedict and Truhon also found infant utterances were
lowest in structured play compared to book reading and free
play, though MLU did not differ across situations. Frequency
of speech and not its complexity appears to change across
situations.

The contrasts between free and structured play indicate
the importance of specifying the purpose of the interaction
in which language samples are taken. Language acquisition
of children, if dependent upon mother speech patterns, must
also be dependent upon the situation that alter mother
speech. Klein (1980) examined language samples of mother-

child speech and father-child speech for the relatively long

27
time period of 6 hours within the infants’ homes. The occur-
rence of "free play" of mothers or fathers to the infants
was relatively rare, and when occurring, lasted for short
periods of a few minutes. Mother-child speech patterns from
free play situations may not be representative of the speech
in typical mother-child interactions.

Summary

Four issues were discussed. First, receptive language
skills were distinguished from productive language skills,
and it was argued that measures of receptive language skills
are more sensitive to the young child's (10 to 25 months of
age) language competence. A second issue viewed arguments
that sensorimotor development is multi-faceted in nature,
with different parts relating to different forms of language
activity. Third, object permanence may be divided into two
periods of development during the sixth stage, the latter of
which may relate to complex language activity. Fourth, the
impact of environmental factors on language development was
described. It was argued that the relationship among cogni-
tion and language may be aided by examining the language
environment.
Statement of_§urpose

Receptive and productive child language are compared to
cognitive development, as measured by object permanence and
means-end relations, and to language environment, as measured
by mothers' speech to their children.

Since highly differentiated object permanence scales

28

hayta'been developed (Corrigan, 1978), a pilot study was under-
‘tak£n1 to develop new stage 6 behaviors for the means-end re-
]jrtions scale of Uzgiris and Hunt (1975). Adding new behav-
iors to the means-end relations scale may add sensitivity to
the scale, especially since only 2 behaviors relevant for
stage 6 activity are found in the current scale. In contrast,
Corrigan's object permanence scaling has :10 separate behav—

iors for stage 6 activity.

Mothers' speech to their children was examined as a pos-

sible environmental factor on child language. The purpose of

the interactions between mother and child was assessed in a
series of videotaped interactions. A wide range of mother—
child interactions typical during the first 2 years of life,
such as feedings, diaper changes, and exploratory play with

different toys and objects were filmed.

Hypotheses

1. Child MLU will not differ across situations, though

frequency of speech may differ across situations, as reported

by Benedict and Truhon (note 1).

2. Object permanence will relate to production of words

representing nonexistence and reccurrence in videotape sit-

uations and vocabularies (receptive and productive). Since
object permanence is theoretically linked to object represen-
tation, object permanence will also positively relate to pro-

duction and understanding of words relating to objects, such

as nouns, proper nouns, and locations.

3. Means-end relations should relate to the number of

29

utterances produced in videotaped situations. Since means-

end relations is theoretically linked to both representa-
tions and verbal requests of goal attainment, these links
should also produce higher correlations to utterances made
in toy and food situations, where specific goals and actions
are present.

4. Child speech development will relate to mother-
speech activity and the child's cognitive development.
Mother-speech is hypothesized as a major source of language

which is being acquired by the children; cognitive develop-

ment as described in hypotheses 3 and 4, is hypothesized as

being linked to early language representations. Three

mother—speech variables are compared, all having theoretical
and empirical (Furrow et al, 1979) relations to child speech.
Furrow et al report yes/no question use (both deleted fron-
tal auxiliary questions, and fully formed questions) will
positively relate to child MLU and auxiliary verb use.
Mother imperative use has been predicted by Furrow et al as
being positively related to child MLU development; impera—
tives are more frequently spoken to children compared to
adults, and this increased frequency marks for Furrow et al
the importance of imperatives to "here and now" language

processing of infants, which relates to later speech acquis-

ition. Newport et a1 (1977) report mother imperative use is
negatively related to child MLU and auxiliary/verb phrase
acquisition; mother yes/no question use is hypothesized as

relating to child auxiliary/verb phrase acquisition. Furrow

30
et al do report mother imperative use at 18 months is nega-
tively related to child auxiliary verb use at 27 months.
Yes/no question use should positively relate to child speech;
imperative use should negatively relate to child auxiliary
verb use; cognitive development should positively relate

and increase the magnitude of the mother to child speech

relationship.

PILOT STUDY

Introduction

The pilot study examined (a) potential new behaviors
representative of sensorimotor stage 6 development for the
concept of space (Piaget, 1954, 1967), as assessed in
Uzgiris and Hunt's (1975) means-end relations scale, (b)
possible alternative scoring for Corrigan's (1978) object
permanence scaling, using errors as done by Jackson, Campos
and Fischer (1978), (0) selection of type and sequence order
of toys to be used in future videotapings of mother-child
language interaction, (d) procedures for interview tech-
niques, language diary administration, comprehension assess-
ment, and Caldwell home inventory scale assessment.
Methods

Subjects Subjects were 17 infants, 9 to 24 months of
age, from two-parent families. Parents were from middle
class occupations or at least one parent had attended college
for two or more years. Five infants were dropped from the
study due to infant uncooperability (p = 2), and inability
to complete all of the cognitive measures (n = 3). The re-

maining 12 infants ranged from 13 to 24 months of age, 6

31

32

males and 6 females.

Procedures Infants were visited in their homes twice
in sessions lasting one to two hours, with sessions separ-
ated by approximately one week. Two infants were seen three
times.

In session one, the object permanence measure, followed
by the means-end relations measure, were administered. Fre-
quent breaks were made, during which questions concerning
the Caldwell home inventory, and toy preferences of the
children, were asked. A set of toys were given to the chil-
dren during a 10- to 15- minute break, approximately one hour
into the first session to observe interest and play with the
toys. The toys were: (a) mechanical car, (b) mechanical
tOp, (0) boy doll, (d) girl doll, (e) foam rubber ball, (f)
yellow 8" wood dowel, (g) multi-color wood block, (h) blue
plastic 2"snuare box. At the end of session one, parents
were given a sample language diary and a checklist, and par-
ents were asked to make entries into each noting difficul-
ties encountered.

Session two continued administration of the object
permanence and means-end relations measures beginning from
the child's highest score in the first session. Frequent
breaks were again given, with breaks used to ask questions
about the diaries, checklists, and Caldwell home inventory.
During one break near the end of session two, the mother
gave three commands to her child, and the child's responses

were noted. The commands were: (a) Get the , (b)

33

Put the on the , (0) Where‘s the .

wordslnmwn to the children, as marked on their checklist,

were:humrted into the blank spaces in the above commands.

Promahues for giving the commands and scoring were developed

by Benedict (1976) and are summarized in Appendix A.

Measures

1. Object permanence. The object permanence scale is

a variation of the standardized Uzgiris and Hunt (1975) scale

developed by Corrigan (1978). Corrigan's scaling adds sev-

eral new "intermediate" hiding situations imbedded within
hiding situations outlined by Uzgiris and Hunt. For example,

the Uzgiris and Hunt scale uses one and three covers in both

visible (object held in hand) and invisible (object held in

container held in hand) displacements. Corrigan adds an in-

termediate step of two covers. Corrigan's scale also records

direction of successful solution of object displacements. The

present scale also records unsuccessful solutions similar to

those reported by Jackson et a1 (1978). For example, an in-

.fant may search at locations A and B, but discontinue search

before reaching target location, C. The object permanence

measure is listed in Appendix B.

2. lWeans—end relations. The means-end relations scale

is rurtzas clearly divisible into new steps as the object per-

manence measure. While the object permanence measure is sim—
ilal‘ert all.levels, the means-end relations measure uses dif-

ferent reaching behaviors, such as pulling pillow, grabbing

string, and holding sticks at different scale steps. Also,

34
the final two steps in the scale do not require reaching,
but instead require some ingenuity and an inhibition of
overt grabbing responses for solution. The first of these
last two steps involves showing the infant a long wooden
bead necklace inside a tall and narrow vase. The infant is
given the necklace and vase separately, and attempts to
place the necklace into the vase are noted. The vase is
sized so that the infant must hold it with one hand and use
the other to insert the necklace. The critical behavior
for success in this task is the wadding up of the necklace
to ease placement. The second of the final two steps in—
volves not placing a filled plastic ring onto a short stick,
after being given several unfilled rings which go easily
over the stick. The addition of new behaviors in the means—
end relations scale will focus on steps that are intermediate
to the necklace and vase situation and the filled/unfilled
ring situation. The last two steps in the scale appear to
require some capacity for representation of unseen goals, a
capacity similar to Piaget's (1954, 1967) descriptions of
sensorimotor stage 6 activity.

Six new situations were presented to the infants requir-
ing a capacity for the representation of unseen goals, in
the sense that overt grasping responses would not aid in
success in the task. The six situations are:

1. The use of a 15" x 10" x 30" "cage" with three

solid sides and a "front" with wooden %" dowels

spaced 6 cm apart. The cage is painted bright

35

blue, and a bright red wooden dowel 8" x %" is
placed inside it. Infants were observed reaching
and attempting to get the dowel. Success required
grasping the dowel at one end and pulling the dowel
out of the cage at a perpendicular angle to the
cage's "bars".

2. The pink and blue puzzle boards from the
Bayley Scales of Infant Development (Bayley, 1969)
were used. Item numbers 137 and 151, completion
and completion after reversal, were recorded.

Item number 155, completion of the blue board in
150 seconds, was recorded.

3. A 4" x 10" x 3" plastic box with six different
sized holes in the top was used. The infant is
given three sticks of different diameters and the
number of sticks placed into the box is recorded.
Each stick was sized to fit into a different num-
ber of holes. One stick fit all holes, a second
stick fit three holes, and the third stick fit

one hole.

4. The child is required to choose a stick with

a curved hook-like end from a variety of sticks

of equal length, in order to reach a distant ob-
ject.

5. The child is required to choose the longer of
three sticks to reach a distant object.

6. The child, seated in a high chair, uses the

36

stick with the curved hook-like end from situa-

tion 4 to reach a target object with a "ring"

that can be snagged by the hook. The child

must ignore two other objects without "rings".

Target and non-target objects were on a box

approximately 2 feet below the child.

3. Language diary. The language diary records the date,
situation, people present and the words understood or spoken
by the child. Procedures and word lists of the language
diary are presented elsewhere (see Benedict, 1976).

4. Language checklist. The language checklist permits
parents to mark words their children know or speak from a
list of frequently acquired words of children under 2 years
of age. The checklist requires understanding or speaking a
particular word in more than two situations before marked as
known by the child. Procedures and words used in the check-
list are presented elsewhere (see Benedict, 1976).

5. Caldwell home inventory. The Caldwell home inventory
is a short standardized scale characterizing the infant home
environment. The Caldwell scale examines mother interactions
with her child and opportunities for the infants to explore
and play.

Subject recruitment and confidentiality

Subjects were recruited by mail, their names obtained
from public record birth announcements from a local paper
from the greater Lansing area. The letters informed poten-

tial subjects of the study assessing cognitive and language

37

development in 1 and 2 year olds. A follow-up phone call
was made approximately 7 days after the letters were sent.
Subjects were then asked for their participation in the
study, and if interest was indicated a home visit was sched-
uled. Subjects were asked for their participation after
discussion of the study's time requirements and purpose
(observing how children play with toys and objects, and how
they can do simple problem solving tasks). All participants
were informed they could leave the study at any time, even
though participation was granted by them.

Participant confidentiality was protected by giving
each subject an identification number. In all data analy-
ses, these identification numbers were used. Access to the
numbers was limited to the experimenter.

Results

The object permanence scoring is listed in table 1.

Ten subjects produced behaviors scaled from the Uzgiris and
Hunt measure. One subject's highest response was scored
from the Corrigan measure, and one subject's highest re-
sponse was scored from the Jackson et al procedure. The
object permanence scale correlated .75 (p ‘<.05) with age.

The means-end relations scoring is listed in table 2.
Two children used the curve-hooked stick horizontally (situ-
ation 6), one child used the same curve—hooked stick verti-
cally (situation 4), and one child chose the longest stick
for reaching (situation 5). The new means-end relations re-

sponses inconsistantly related to the Uzgiris and Hunt scale

38

.mnpsoe Cw zeogm ow<n

 

 

 

 

was neon m manama>ca weave; Hm mHNEme ma.sa
no» onCHm m manfimfi>cfl pamsoomammao mm mausofl oo.wa
mm» neon m manawa>ca pamEmomammae Hm mango“ ma.om
no» neon m oHpHmH>cH pamsmomamwwp on madam“ om.mm
mm» neon m manama>cfl pamEmomHQmae on mHmEmm om.mm
no» oawzflm m oHpHmfl>nfl PQmEoomameU mm onEom oo.:m
o: mawcflm m manama> pcmsmomammflc HH mama mm.mfi
mma meCHm m manwwfl>cfl vaEmomHQmwp mm mama oo.mH
mm» neon N manama>ca passwomammae mm mama om.mfi
mm» 5909 m manwmfi>Ca psosmooammao mm mane mm.mm
mom mawcflm m manflmfi>cfl pCmEoomHmwfiv mm wane om.mm
we» oawcflm m manflww>cfl PcoEoomamwao mm mane oo.mm
mmmmmmm mcowpoohflo macapmooq MMMHMS Paosmomanwwm mwmamm mmm..mmm¢
condom nopoom no wcfiowm manom

 

 

H magma

mocMEpohpom monocmshoa poonno

.m:v:oe CH csozm mw<m

 

39

 

soapm mcfim: Scoop mCHpcmpxo m oHoth mm.:fi
omm> can momaxom: m cause“ oo.mH
xoflpm wcfim: canon waUCmpxo m mamsom mm.om
mop pow 0» wappm Havapuo> mcflaasm o madam“ om.mm
mwcao umaaaecs\emaaae m mHmEme om.mm
mop pow op wsfinpm Hopcosauo: wcaaazm m mango“ oo.:m
Boaaam so how pow 0p soaafim mowaasm m mama mm.mH
omm> can oomaxom: w mane oo.mfi
weaken wean: scoop wcflUCmpxm 5 came om.wH
mmn> paw mooaxom: m mama mm.mm
hop pom op wcflppm Hmpcouflhon wcwaasm m mans om.mm
hop pow op mcfippm Hmowppm> mowaasm m mama oo.mm
poa>mnmm mo soapmflpomom MMMHflZ. WNW mmmd

manom

 

 

mocmshompom moowwmamp vConCMmS

N manna

40

steps of the necklace and rings. One child used the hooked
stick horizontally (situation 6) but did not wad the wooden
necklace beads to facilitate vase placement. Another child
did not wad the necklace but could choose the longest stick
for reaching (situation 5).

Successes for situation 1, 2 and 3 appeared more con-
sistantly and frequently (responses = 3, 5, 4) compared to
situations 4, 5, and 6 (responses = 1, 1, 2). However, two
children who completed the red board (situation 2) did not
solve the necklace situation. Also, one of the two child-
ren completing the blue board (situation 2) did not succeed
in the necklace problem. The standardized means—end rela-
tions scale (Uzgiris & Hunt, 1975) did not correlate with
age (r = .19, p < .10) and did correlate with object perman-
ence (; = .65, p < .05).

Toy preference was noted during a 5 minute free-play
period during the first home visit. Preferences were re—
corded by marking which toy was being handled out of a set
of toys given to the infant. Play frequencies are listed
in table 3. More than one toy could be counted during each
10 second observation period. For example, simultaneous
play with boy and girl dolls occurred frequently, as did
play with stick and ball. Extended play occurred frequently
for "movable" toys, the two "mechanical" toys, and the dolls
with movable appendages and clothing, while less "movable"
toys were ignored, with the exception of the foam rubber

ball.

41

Table 3

Toy preferences during a five minute free play

 

 

 

22y Frequencya
Car 117
Top 70
Male doll 55
Female doll 48
Ball 45
Stick 21
Block 18
Plastic box 8
No toy play 63

 

aObservations were recorded once every 10 sec-
onds during the free play; more than 1 toy
could be counted during each observation. All

toys were presented as a group to each infant.

42

Discussion

Though object permanence behaviors observed here were
consistant with the Uzgiris and Hunt (1975) measure, the
intermediate steps described by Corrigan (1978) and Jackson
et al (1978) were also observed, although not as frequently.
A combined approach, using Corrigan's and Jackson et al's
procedures with the Uzgiris and Hunt scaling appears appro-
priate for future use.

Several inconsistencies were found for the new means-
end relations behaviors, especially situations 4, 5, 6 and
situation 2's blue board. Situations 1, 3, and situation
2's red board have some consistency to the standardized
Uzgiris and Hunt scale steps, and their tentative use seems
appropriate.

Toy preference appears related to toy complexity, in as
much as some toys contained movable parts. Movable toys,
such as the toy car and top, elicted the most attention and
play from infants. Social toys, such as male and female
dolls, also elicited a great amount of attention. Non—mov-
able toys were largely ignored. Infant agitation, crying
and distress were more frequent at play session termination
when the infants were playing with movable and social toys.
Movable toys seemed to produce the most distress at play

termination.

STUDY II

Introduction

This study examined infant language, both receptive and
productive, as a function of: (a) object permanence,
(b) means-end relations, (0) mother to infant speech across
feeding, diapering and play situations, (d) infant environ-
ment as measured by the Caldwell home inventory. Modified
versions of the object permanence and means—end relations
scales of Uzgiris and Hunt (1975) as described in the pilot
study were used. Language measures used parent checklists
and diaries described in the pilot study. Infant speech
was recorded on audio and video tapes. This redundancy
of infant language measures helped to accurately character-
ize infant language competence.
Methods

Subjects Subjects were 30 first born white middle class
children (15 male, 15 female) and their parents. Three age
groups (14, 18, & 22 months, p = 10) were cross-sectionally
examined. Only families with mothers working fewer than 30
hours a week were recruited because of the time requirements
needed to maintain the language diaries. Middle class
status was defined by Hollingshead's (1971) classifications
2, 3, and 4, though college families were also recruited

(minimum of 2 years of college for at least 1 of the parents).

43

44

White middle class infants were recruited to limit possible
social class effects from confounding results, and to en-
hance this study's comparability to other studies which
have emphasized white middle class subjects.

After the start of data collection, 4 infants not in-
cluded in the 30 infants mentioned, were dropped from the
study. One mother asked to be dropped from the study, 2
families were dropped because of illness, the father in 1
case and child in the other. One family was dropped when it
was revealed that the father's 2 years of college experience
was for a G.E.D., since the father did not finish high
school. This last family also did not belong to Hollingshead
classifications 2, 3, or 4.

Subjgct recruitment and confidentiality The procedures
for subject recruitment and confidentiality followed those
described in the pilot study, except for the subject's first
names which were not coded. First names were frequently
mentioned by mothers in the tapes. Last names were removed
from the transcripts, since these were much less frequent.
Subjects were often referred to by their first names as well
as by their code number. Parents were informed of this pro-
cedure before consenting to participate.

Procedures Each measure was administered in the in-
fants' homes, in 5 sessions lasting approximately 50 to 90
minutes apiece. Four families were visited 6 times, and one
family was seen 7 times. Home visits were scheduled at the

parents' convenience, during hours infants were likely alert

45
and awake. The home is an infant's most familiar environ-
ment, likely increasing exploratory activities to the study's
tasks, toys and situations, and decreasing non—study ex-
plorations found in unfamiliar environments such as a class-
room. Home visits also facilitated observing specific in-
fant language behaviors, such saying "chair" to one particu-
lar chair in the child's home. Two observers were present
on most home visits. Observer 1 administered the cognitive,
language, and environmental measures. Observer 2 recorded
the data.

Language diaries were maintained for at least 3 weeks,
beginning with the first visit. There were 2 diaries, 1 for
comprehension and 1 for production. The parents were also
given a checklist for their children's receptive and produc-
tive language on the first visit. Parents were instructed
to mark the checklist within 3 days, and to add any new
language activity to the diaries.

General procedures for sessions The first session in-
volved introducing the mother to her role in the study, a
role consisting of keeping a diary and checklist of her
child's language during her study participation. Length of
participation was determined by the scheduling of the home
visits, though all diaries were maintained for at least 3
weeks. Mothers also uttered short commands to their chil-
dren to assess receptive language, and were also filmed in
videotaped sessions. Each.mother was interviewed for the

Caldwell home inventory, and was present during presentation

46

of the object permanence and means-end relations measures.
The relatively large role that the mothers played in this
study was intended to optimize language and cognitive per-
formance by her child.

After the mother agreed to participate, she was shown
the diary and checklist procedures. Four future sessions
were scheduled at the end of the first session, though most
mothers scheduled only the second session.

The second session examined performance on the object
permanence and means-end relations scales. The mother's
diary and word checklists were also reviewed, and mothers
described circumstances under which their children had
understood or had spoken words. Special care was taken to
ask about words relating to locations, object features,
possessives, non existence, and recurrance. Words recorded
ambiguously, or in situations containing non-verbal cues
(gestures), were dropped from the diary and future inclusion
in the child's vocabulary.

Language comprehension was examined in the third and
fourth sessions. Mothers read 21 stock sentences, contain-
ing 3 to 5 words apiece, to their children. The sentences
contrast knowledge of possessives, locations, and relations
between objects as developed by Benedict (1976). Words
taken from the children's diaries were used in the sentences.
Each sentence was repeated up to 5 times. (see Appendix B
for list of sentences, scoring and procedures).

Toward the end of session 4, object permanence and

47

means-end relations measures were administered a second time.
Each child started at the highest step solved in each meas-
ure in the first administration, and continued until 3 suc-
cessive failures at a lower, same or higher level were
reached.

The fifth visit involved videotaping of mother-infant
interactions with toys. Videotapings of diaperings and
feedings were made during the third, fourth, and fifth
visits, depending on the spontaneous needs of the children.
Audio recordings of the children were made during the last
3 home visits in free play situations. The free play situ-
ations lasted approximately 10 minutes, and provided esti-
mates of child speech from a longer time period than the
videotapes. Mothers were usually interviewed with the
Caldwell home inventory questions during these free play
sessions. Mothers also used this time as a break from data
collection for coffee or refreshments.

Measures Many measures of the present study were used
or reviewed in the pilot study. Measures of the present
study are:

1. Object permanence

2. Means-end relations

3. Parent checklists and diaries for both

comprehension and production

4. Caldwell home inventory

5. Language comprehension procedures (see Appendix B)

6. Videotaping of 6 toy exploration situations,

48
a diapering and a feeding of a snack

7. Audiotaping of free play situations

Videotapingfprocedures All videotaping was filmed in
the infant's home, in rooms appropriate for each situation
and which were comfortable for the mother and infant. There
were 8 situations in the study, divided into 5 groups (3
types of toys, diapering and feeding). Each situation was
filmed for at least 90 seconds. The duration of a situation
depended on both participants, though mothers primarily stop-
ped the taping; general reasons for stopping were: (a) moth-
er lost interest in showing a toy, (b) mothers perceived
their infants as having lost interest in the interaction,
(c) the interaction was completed. Maximum time for any
interaction was set at 180 seconds.

Six situations filmed infant play with objects and
mothers showing their infants the objects. The objects
and their presentation order were determined by the pilot
study results. Because of time restrictions, each toy was
taken from the child shortly after 180 seconds, and replaced
with the next toy. This was usually no problem when the
child was no longer interested in the toy, though some in-
fants regained interest when replacement time arrived. The
toys were therefore ranked in terms of interest by the pilot
study results, and the presentation order was reversed, so
that the least interestng toy was presented first, and each
successive toy became more interesting. It was believed this

procedure reduced disruption in this study from crying,

49

protesting infants.

The remaining situations, diapering and feeding, were
filmed when feedings or diaperings naturally occurred. Due
to the private nature of diapering and feeding, mothers were
informed that these situations were optional.

The videotape equipment was set up in the living rooms
(or room where administration of the measures occurred) for
sessions 1,3,4, and 5 to reduce infant fear or curiosity
during the actual taping. The videotape equipment was port—
able which permitted the videotaping of spontaneous feedings
or diaperings in room(s) appropriate to that activity.

Mothers were informed of the videotapings during the
first home visit, and they were told any taping (audio or
video) could be stopped at any time, even in the middle of
taping. All taping was discontinued when mothers received
phone calls or visits. Each situation is described below.

1. The diapering videotaping was spontaeous. When
the infant informed his/her mother of a diapering need, and
that need confirmed, the mother then informed the experi-
menters. The experimenters then carried the videotape
equipment to the diapering room. Videotaping began when
the child was first placed on the diapering surface.

2. The feeding was more of a snack than a meal, con-
sisting of milk (or juice or water) and cookies (or fruit
or crackers or a sandwich), depending on the preferences of
the mother and child. The feeding was taped in a location

used for snacking, and in the usual manner for each child,

50

such as sitting in a highchair or adult chair, or standing
up and free to run around. Videotaping began when mothers
started to prepare or reached for food. Any child assist-
ance in food preparation was recorded. Some children were
asked to select their snack as a normal part of their snack-
ing routine. These questions about food preferences were
videotaped.

3. Toy videotapings were filmed in the fifth visit.
Six toys were filmed. Each toy defined a new ”situation”,
which could be used as a new conversational "focus" by the
mothers or children. There were 2 object toys, 2 social
toys, and 2 action toys. Short breaks were taken after each
toy, with a longer break (2 to 5 minutes) occurring after
the third toy. Videotaping began when mothers unwrapped a
toy in front of their children from a small cloth blanket.
Toy situations were taped in a room frequently used by a
child for play. Before taping, the child's toys were re-
moved from the room, usually before the start of the session
to prevent interference with the study's toys. Video equip-
ment was set up approximately 8 feet from the interaction,
and the camera followed the child moving about the room.
Mothers determined seating arrangements with her child (face
to face, both facing toy, infant on lap, etc.). The follow-
ing toys were videotaped in the order presented:

1. multi-colored stick

2. foam rubber ball

3. female doll

51

4. male doll

5. push-pull top

6. spring powered friction car

The diapering and feeding situations were spontaneous
and no special instructions were needed prior to taping.
Since the toy situations were structured, mothers were
given the following instructions,

I want you to sit with your child in a way that's

comfortable for both of you. Remember, I will be

handing you toys and objects to show to (child's

name), so please sit in a way that's also comfort-

able for you to show the toys and objects.

Once seated comfortably, mothers were also given the
following instructions,

I am going to hand you several objects, one at a

time. I would like you to show (child's name)

something about each object. You may show your

child as many or as few things about each object

as you would like or as much as you think your

child would like to know about each object.

Please hand me each object after you're finished

showing it. You can stop showing the object

any time you think your child is no longer inter-

ested in the object. You can also stop any time

you are no longer interested in describing the

object, though I would like you to try to de-

scribe the object for at least one minute. Do

you have any questions?

If no questions were asked, videotaping was started.

Data reduction and coding Data reduction and coding
was made for mother and child speech in videotaped situa-
tions, and child speech in the audiotaped free play situa-
tions. All mother speech was categorized twice, once for
lexical speech and once for syntactical speech, as done by

Furrow et al (1979) These categories are described below.

52

Lexical categories

1.
2.
3.

\O(D\To\

10.

11.

12.
13.

words: get, ball, Sandy, on.

sentences: Get the ball. Put it in here.
nominals

nouns: Jeffrey, ball, orange, table.
personal pronouns: you, I, me, they.
objective pronouns: this, that, one, there.
verbs

main verbs: eat, get, pull, put, go.
auxiliaries: should, shall, can, will, would.

copulas: be, is, seems, appears.

. modifiers

noun modifiers: big, little, broken.
verb modifiers: (bring) up, allmost, (come) here.
tense: This category codes future and past tense.
plurals: Both plural nouns and pronouns are coded.
genitives: my, daddy's, your, mine, his.
negatives: no, not, (do)n't, (ca)n't.
contractions: Categories code contractions in
copulas, auxiliaries, and negatives.
prepositions: All prepositions were coded, in—
cluding prepostions without objects.
noun/verb ratio: Main verbs were used in this ratio.
MLU: Mean length of utterance was coded as described

by Brown (1973).

Syntactical categories

1.

declarative sentences: That is a ball. It's raining.

53

2. imperative sentences: You get the ball. John look

at this.

3. questions

yes/no inversions: Can you do that? Do you like it?
Wh- questions: Where's the ball? What are you doing?
tag questions: You can see it, can't you? It's
pretty, isn't it?
yes/no other questions: You wanna get down? See the
car?

4. interjections and fragments: and, well, the blue car.

Since different videotaped interactions contrast social
and object oriented activities (dolls versus cars), the pro-
noun category in lexical coding was divided into personal
and objective pronouns.

Fragments, coded within the syntactic categories, utter-
ed with a rising intonation pattern (e.g. "No doll?") were
counted as "yes/no other" type questions. Essentially com-
plete sentences produced with a conjuction to start the ut-
terance were counted as complete sentences (e.g. And there
is the new dolly).

Data reduction and coding for child speech (videotaped
and free play situations) were categorized for MLU, verbs/
utterance, nouns/utterance, auxiliaries/verb phrase (Furrow
et al., 1979; Newport et al., 1977), words coding non exis-
tence and recurrence (Corrigan, 1978), and verbal request
making.

Child language recorded in diaries and checklists were

54

categorized into 15 word classes for both receptive and pro-

ductive

language (Benedict, 1976; Zachery, 1978). These

word classes and some examples are presented below.

Vocabulary word classes

1.

9.

10.
11.
12.
13.
14.

15.

verbs: get, bring, do, where.

features: big, little, broken.

articles: a, the.

colors: blue, pink, yellow.
recurrence: more, again.
possessives: mine, yours, daddy's.

locatives: up, in, on. Locatives were coded in

reference to self and to other objects.

negatives: no, not, (do)n't.
positives: yes, OK.

non existence: allgone, all done.

amounts: some, all.

numbers: one, two, three.

nouns: table, chair, ball.

proper nouns: Jimmy, Fred, Billy.

total words: This category sums all words known

by the infant.

Locative terms coded whether the words were referenced

in relation to the child, to other objects, or to both forms

(Pick, Yonas, & Rieser, 1979).

A child's general level of word comprehension was as-

sessed in home visits 3 and 4, during which mothers read 21

stock sentences to their children. Each sentence was read

55

up to 5 times apiece. The maximum number of words responded
to by the children in any one sentence reading was recorded;
comprehension of relational meanings was noted. Relational
comprehension was scored with successful discrimination of
several modifying adjectives embedded within the sentences.
For example, a child's correct selection of a big ball and

a little ball, in successive sentences, show the discrimin—
ation of the modifying word pair big/little. Also, relational
comprehension was scored with an understanding of possessives
across sentences, such as correctly selecting articles of
clothing (shoe or sock) belonging to the child, child's
mother, and child's father in 3 different sentences.

During the breaks in sessions 3 through 5, the chil-
dren's vocabularies (comprehension and production) from their
checklists and diaries were examined. Mothers reconstructed
situations under which language activity occurred. For pro-
duction, mothers were asked whether objects were present or
actions ongoing, during the child's speech. Questions were
asked on the production of modifiers, and abstract semantic
categories, such as possessives, features, amounts, non-
existence, and reccurrence. Modifiers and abstract words,
produced in situations with the object present or actions
ongoing, were noted but not placed into the children's vocab-
ularies. Words were included in a child's vocabulary when
produced in more than two different situations. These sel-
ection procedures are consistent with those of Corrigan

(1978), and these procedures eliminate words associated with

56

a particular context and modifying/abstract words associated
with particular objects.

Mothers were also asked to reconstruct circumstances
under which their children understood words marked in the
children's diary and checklist. Words which were under—
stood with the help of gestures were eliminated from the
children's vocabularies. As in production, placement of a
modifier or abstract word into a child's vocabulary required
consistent comprehension of the word in two or more situa-
tions, and for abstract words (recurrence and nonexistence),
the objects must have been out of view or the action not
ongoing.

The children's productive vocabularies were supple-
mented with words produced in the video and audio tapings.
The children's comprehension vocabularies were supplemented
with child word comprehension behaviors from the sentence

task of sessions 3 and 4.

RESULTS
Introduction

Transcript and coder consistancy for the video and
audio tapes are examined. Percent agreement, which is more
conservative than correlations, was calculated to estimate
this consistency. The relations of the new means-end re-
lations behaviors to the standardized Uzgiris and Hunt
(1975) scale are also examined.

A series of analyses of variance contrasting sex, age,
videotaped situation to child attention, MLU, nouns/utter-
ance, verbs/utterance and auxiliaries/verb phrase are pre—
sented. Next, partial correlations, with age variance re-
moved are presented; these correlations relate means-end
relations and object permanence to the children's vocabul-
aries (receptive and productive) and child speech from the
audio and video tapes.

Relations between mother and child speech are examined
next, comparing different videotaped situations. Finally,
three-way relationships between mother-speech and cognitive
development are examined in series contingency tables and
multiple regression equations. Discussion of these results
is presented in the next section.

Transcript Coding and Consistengy

The videotapes and audiotapes were transcribed by four

57

58

research assistants. Every transcript was reviewed twice by
the principle investigator for errors and ommissions.

Transcripts were coded for mother and child language
categories by the primary investigator. As a check, 6
transcripts were completely coded for all categories by 2
independent raters. The 6 transcripts represented each sex
by age cell in the study (male and female x 14, 18, & 22
months olds). Within cells, a transcript was randomly sel-
ected for the check by the roll of a die.

All coders were trained with a videotape of a mother
playing with her 24 month old in their living room. Activi-
ties in the videotape included playing ball, book reading,
and watching through a large picture window a dog running
around outside the house. The training tape was approxi-
mately 45 minutes long.

Training tape transcript coding produced a percent
agreement of 84% for rater 1, and 91% for rater 2 to the
training tape coding of the principle investigator. All
discrepancies were resolved through discussion.

Coding of the 6 videotaped transcripts produced percent
agreements ranging from a low of 87% to a high of 93% to the
codings of the principle investigator. Averaged percent
agreement across all 6 transcripts was 90.1%.

Means-end relations scale consistency

Some means-end relations behaviors were infrequent (see

Appendix C for raw data). Only 1 infant was able to remove

the stick from the cage within 3 tries without getting

59
the stick stuck on the bars of the cage (Task #5). Because
of the low frequency of this behavior, this situation was
dropped from further analysis. The two remaining "new" sit-
uations show some discrepencies to performance on the stan-
dardized situations of the necklace and the rings. The
sticks and the plastic box situation (#6) showed 10 suc—
cesses, 8 successes accompanied by failures in the necklace
situation, though 2 of these had successes in the rings sit-
uation. Since the Uzgiris and Hunt (1975) scaling proce»
dures record the infant's highest level of scale behavior,
these latter 2 children were considered consistent with the
scaling, reducing the discrepencies to 6 out of 10. Six out
of ten discrepencies, however, appear very high so this sit-
uation was dropped from further analysis.

The red board from the Bayley Scales of Infant Develop-
ment (#7) showed three successes, two of which were accom-
panied by failures in the necklace situation, though one of
the failures was successful in the ring situation. Thus,
discrepancies for the red board situation were one out of
three. However, the discrepancy for this one child was se-
vere since his next highest scoring behavior was horizontal
string use, which is three steps below the minimum stage 6
situation of the necklace. Because of the severity.of the
discrepancy and that it represents a 33% error rate, and the
low success rate, the red board situation was dropped from
further analysis. Thus, the means-end relations scale used

in this study was the standardized scale of Uzgiris and

60

Hunt (1975).

Examining the data for the standardized ring and neck-
lace situations shows two children succeeding in the ring
situation but failing in the necklace situation. This in-
consistency represents a 50% error rate for the ring situ-
ation. Uzgiris and Hunt code a child's highest performance
in a scale, and so this inconsistancy was ignored for future
analysis. That is, those succeeding at rings were counted
as achieving the highest level in means-end relations. This
approach is different from the analysis of the "exploratory"
means—end relations behaviors, however a conservative ap-
proach was considered necessary for exploratory steps.
Standardized coding procedures do report inconsistancies in
the successful solving of the ring and necklace situations.

Uzgiris and Hunt (1975) report the inter—session stab-
ility average for the standardized means-end relations scale
behaviors was 75.5%, and this stability averaged 56.9% for
the necklace situation and 88.2% for the ring situation.
These percentages indicate the probability of a child re-
peating a successful performance in a step within the same
situation. The lower this stability percentage for a step,
the greater the chance a child may fail one step and pass a
higher step in a scale. Uzgiris and Hunt's scalogram ana-
lysis shows mean-end relations as having the second lowest

index of reproducibility (consistancy index = .81).

61

Hypothesis One

The first hypothesis analyzed child speech in the vid-
eotaped situations in a series of repeated measures analysis
of variance.

A 2 x 3 x 6 (sex x age x situation) analysis of vari-
ance with situations treated as a repeated factor was calcu-
lated on the duration of each videotaped toy situation.

This analysis compared the extent the children continued
their play; since the 6 toys were presented in a fixed order,
toys presented last may have been neglected due to fatigue.

Though duration of the videotaped situations may not
perfectly correlate to actual infant attention to the toy of
that situation, it is an index of the mother's perception of
her child's interest in the toy. Occasionally, mothers re-
ported stopping the interaction as they had run out of
things to say about the toy.

The duration of the situation analysis indicates that
attending differed as a function of the situation, F_(5,870)=
8.48, p <.01. Trend analysis indicates that the linear
trend accounted for 95% of the variance due to the situations
factor. There were no other significant main effects, trends
or interactions. Thus children increased the amount of time
playing with the toys as the number of situations increased.
This finding confirms the results of the prestudy in which
static, non-moving toys were least interesting, and dolls
and action toys were most interesting to the children.

Scores for utterances, MLU, nouns per utterance, verbs

62

per utterance, and auxiliaries per verb phrase were compared

in a series of 2 x 3 x 8 (sex x age x situation) analyses of

variance (situations treated as a repeated factor). The num-
ber of utterances spoken by the children showed a significant
sex x situation interaction, §_(7, 1218) = 2.19, p‘<.05,

and significant effects for age, F_(2, 174) = 5.73, p < .05,

and situation. E (7. 1218) = 3.05, p < .05. Cell means show

the interaction was due to the larger number of utterances

by females, and the attenuation of utterances in the diaper-

ing, toy top and car situations for all subjects.

Only age was significant in the MLU analysis, F (2, 174)
= 8.73, p <.05, indicating speech increases in complexity
with age. The verb/utterance analysis contained no signifi—
cant main effects or interactions, though a marginal sex x
situation interaction emerged. E (7. 1218) = 1.97. P ‘ .06.
Cell means show males produced more verbs in toy top, diaper-
ing and toy car situations, while females produced more verbs
in feeding and toy block situations. The noun/utterance analy-
sis showed significant main effects for age, F (2, 174) = 2.88,
p <.05, and situation, F (7, 1218) = 2.31, p <.05. A poster-
iori comparisons indicate: (a) 22 and 18 month olds produced
more nouns/utterance than 14 month olds and that they did not
differ from each other, and (b) fewer nouns were produced in
the toy top situation than any other situation.

The auxiliaries/verb phrase analysis indicates a signifi-
cant main effect for age. Examination of cell means show

that no 18— or 14-month old produced any auxiliaries/verb

63
phrase.

In sum, language complexity (MLU) did not change as a
function of situation, though the frequency of speech did
change. Children spoke less in the moving toy and diapering
situations; girls tended to speak more than boys. Boys
spoke more verbs when playing with the moving toys and dia-
pering; girls spoke more verbs when playing with the block
and during feeding. The children used verbs to describe on-
going activities, such as "Car mov(ing)" and "Eat now". The
ability to speak nouns and auxiliary verbs is related to age.
The present data supports hypothesis 1, MLU did not change
across situations, frequency of utterances and verbs/utter-
ance did change across situations; these changes in speech
activity may alter the relations of mother-speech to the
children across situations.

Hypothesis Two

The second hypothesis examines the relationship of ob-
ject permanence performance to child speech in videotaped
situations, receptive and productive vocabularies. The 6
toy situations were combined into 3 general categories:

(a) non-moving toys consisting of the stick and ball,

(b) social toys consisting of the male and female dolls, and
(c) moving toys consisting of the top and car. Feeding and
diapering situations were also analyzed.

Table 4 lists partial correlations of object permanence
to child vocabulary categories. Table 5 list partial corre-

lations of object permanence to child speech in the video

64
Table 4

Partial correlations of object permanence performance to

number of words in child vocabulary categoriesa

 

 

object permanence performance

 

 

 

 

vocabulary_categopy comprehension production
Total words .15 .26*
Nouns .23 .16
Proper nouns .13 .04
Features .33** .20
Colors .13 .06
Person-object locatives .31** .05
Object-object locatives .06 -.06
Amounts .15 .15
Possessives .22 -.16
Nonexistence .08 -.02
Recurrence .35** .08
Verbs .18 .04
Sentence comprehension .08 ---
Relational comprehension .19 ---

 

Note. Uni-directional tests of significance were used since

 

positive correlations were predicted in hypothesis 1.
aAge effects are removed from these correlations.
* p < .10

**p < .05

65

mo. v a:
Ca. V Q...
wCOMPmaoh ommnp thh ©o>oEmh ohm mpommmm om¢m

.H mamonpoman ow oopowoonm who mCOHpmaon o>flpwmom mocam .mGoapm

 

:Hohnoo Hoappom omonp onmasoo op cows who cocooflmflcwflm Ho mpmop HonoweoonfionfiCD .opoz

 

 

 

 

 

 

 

HH. mo.- mo.- ofi. om. oH. mzop woa>oz

no. so. Ho. ofi.- ma. ao. asap Hmaoom

0H. NH. Ho. wH.: mo.: 00.: maop wcw>oE::oz

no. HH. :0. mo.: mo.: a0. wcflooom

NH. NH. ma. mm. **©m. Hm. mcflnommflm

0H. wH.i **wn. mo. No.i NH. hoamloopm
mesosoPMVm ommnnmrnpo> monopopps oocmnoppb mma moocmnoppo oocoEnomuom
Homsoo \mofinMwHHxs¢ \wcsoz \mnno> monocospom
pomfino

 

 

omCoHPMSPfim ooQMPooofl> one

zmamnoonm CM nooomm pawno op monocwspom Poowno mo mcowpmaohhoo Hmfippom

m manna

66

taped and free play situations. Object permanence relates
to nouns/utterance in free play (g = +.38, p < .05), MLU in
diapering (g = .31, p < .05), though not to object locatives
(p = .08, p < .05).

Partial correlations, controlling for age, were calcu—
lated for object permanence to: (a) words coding recurrence
in freeplay, diapering, feeding, and social situations, and
(b) words coding nonexistence in freeplay. Correlations
were not calculated for the remaining situations since not
enough words coding recurrence or nonexistence were pro-
duced in those situations. The partial correlations of re-
currence and nonexistence to object permanence were low
(range -.20 to +.24, p .<.10), suggesting little relation
between these categories and object permanence once age fac-
tors are removed.

In sum, hypothesis 2 is partially confirmed; object
permanence is related to some object-defined categories
though these relations are mainly in receptive vocabulary.
Other object related categories were not related to object
permanence (nonexistence). Object permanence does not ap-
pear to relate to increased child speech, except for MLU in
diapering and nouns/utterance in freeplay; object permanence
relates more to vocabulary build-up.

Hypothesis Three

The third hypothesis examines the relationship of means-

end relations to child speech in videotaped situations, re-

ceptive vocabularies, and productive vocabularies. As in

67
hypothesis 2, the 6 toy situations were combined into 3 gen-
eral categories: (a) non-moving toys, (b) social toys, and
(c) moving toys. Feeding and diapering situations were also
analyzed.

Table 6 lists partial correlations of means-end rela-
tions to child vocabulary (receptive and productive) cate-
gories. Table 7 lists partial correlations of means-end re-
lations to child speech in the videotaped and free play sit—
uations.

Generally, performance on the means-end relations scale
negatively correlated (age variance removed) to production
vocabulary categories, though only the nonexistence (in pro-
duction) category reached statistical significance (r = -.39,
p < .05, two-tailed test), and possessives (in production)
were marginally negatively related (r = -.33, p < .1, two-
tailed test).

Means-end relations negatively related to verbs/utter-
ance in free play, diapering, feeding, and moving toys situ-
ations (pg = -.55, -.42, -.39, -.38, p < .05); auxiliaries/
verb phrase negatively related in feeding (r = -.52, p < .05)
Causality statements marginally related to means-end rela-
tions in social and diapering situations (rS = .29, .28,

I) <.10). Utterances marginally related to means-end rela-
tions in the social situation(r = .26, p < .10).

In sum, hypothesis 3 is weakly confirmed since causality

statements marginally relate to means-end relations in 2 sit-

uations, social play and diapering; a marginal relationship

68
Table 6

Partial correlations of means-end relations performance

to child vocabulary categoriesa

 

 

means-end relations performance

 

 

 

 

vocabulary categories comprehension production
Total words .06 .27
Nouns .28 .11
Proper nouns .29 .16
Features .01 -.24
Colors —.11 .02
Person-object locatives .08 -.18
Object-object locatives -.16 -.11
Amounts -.18 -.14
Possessives -.10 -.33*
Nonexistence -.06 -.39**
Recurrence .22 -.02
Verbs .13 -.14
Sentence comprehension -.09 -—-
Relational comprehension .03 ---

 

NQEQ. Bi-directional tests of significance were used to com-
pare these partial correlations.

aAge effects are removed from these relations.

*p < .10

*‘I’p < .05

69

mo. v o**
Ca. v m*
.wCoapoHon owonp Sony oo>osop who mpoommo ow¢m
.oocmoamwzmam mo memop Ansofipoonwoufln om: mooflpmaon pocpo Ham “m mwmonpoman

CH oopowooum who mCoapoHoh o>flpflmoa oocfim .mpCoEmpmpm Homsoo pan moocmnopps Ho

 

mowuowopoo nooomm oaflno ounQEoo op now: who cocoowwflcmflm “o mpmop Honoflpoohaonacb .opoz

 

 

 

 

 

 

 

HH. *mm.: **mm.: wo.: HN.: do. whov ma>oS
*mm. mm. HH.- om.: mo. *om. maop Hmaoom
3N. *Hm. HN.: 0H. Ho.: ma. mcfl>oEICOz
30.: Nm.| **mm.l 0H. OH.I 00.; MQHUmom
*mm. 30.: **N:.: no. *dm.i 3H. mcfluommflo
Hm. mo.- **mm.- om.- 00.- ma. amaonmmom
mpCoEopMpm omonnmrnpo> moocmnopp: mooconoppb .mma mooconmppb oocosnownom
Homomo \moflnmflafixs< \mczoz \mppo> mCOfipmaom

ozonmcooz

 

 

.mmCOMPmSPHm oomwpooofi>

ocw awamioonM Ca gooomm oafino op mCOHPmaon ozoimcoos Mo mCOflpmaonnoo Hmflpnmm

m oHQmB

70

also occurred in 1 situation, social play. Means-end rela-
tions negatively relate to verbs/utterance in free play,
diapering, feeding, and moving toys situations; the last 3
situations were examined in the verbs/utterance analysis of
variance in hypothesis 1 which showed increased verb use in
moving toys and diapering situations by males, and in the
feeding situation by females. Since only 4 subjects achieved
stage 6 in means—end relations, the negative relations can
be attributed to increased verb use in the situations cited
rather than to a large number of stage 6 children producing
few verbs. Increased verb use may assist protracted goal
attainment which may be needed by stage 5 children in their
reaching and tool use activities.
Hypothesis Four

The relationship of mother-speech and child cognitive
development to child speech is explored in a series of par-
tial correlations and multiple regressions; relationships
are calculated for mother and child speech produced within
each of the 5 major videotaped situations of social toys,
non-moving toys, moving toys, diapering and feeding. Mother
speech is examined in 2 categories: (a) sentence categories,
and (b) grammatical categories. Child speech is examined by
6 variables: (a) MLU, (b) utterances, (c) verbs/utterance,
(d) nouns/utterance, (e) auxiliaries/verb phrase, and (f)
causality statements. These partial correlations are pre-
sented in tables 8, 9, and 10.

The partial correlations do not show a statistically

71

mo. v Q*

muofipmaopnoo omonp Bonk oo>oEon who mpoowwo om<m

 

 

 

 

 

*0m.- *0m.- Hm.- *0:.- m0.- 00. ma. m0.-
0m.- m0.- mm.- 0H.- *0m.- mm.- Ha. 0H.
0m.- Hm.- mm.- *m:.- mo. 0H.- 00. ma.-
m0.- m0.- 0H.- Ha.- 0H. HH. :0. Ha.-
sm.- :m.- m0.- an. 00. m0.- 0m. m0.
sm.- mH.- *ms.- *H:.- :0. H0.- mm. NH.
N0.- mm.- mm.- 0H.- mm.- 00.- N0. 0m.-
00.- ma. RH.- 0H.- 0H.- 00.- HH. 2H.
0H. *ss. nfi.- 0H.- :m.- H0.- ma. on.
ma. mm. HH. ma.- sa.- HH.- *mo. *mm.
Hm.- 00. mm.- 0m.- mm. AH. 0m.- 00.-
mm. mm. 00. mfi.- 0m. ma. on. m0.-
m0.- mm. 0H.- HH.- 0H.- 0m.- :0.- 0H.-
mm. :m. HH.- 00. 00. H0. mm. ma.

amm.- *ms.- 0H.- *sm.- m0.- 00.- Ha. Ha.-
mm.- HH.- HH.- NH.- HH.- 0H.- 0m. no.

wMOP wcmwwz mMOP Hmmmww waow mcfi>wm£Coz Dmcfioowws

m0. mm.- unmanaeoe ppm>
Hm.- ma. mpmaoaeoe :soz
:0.- 0N.- mnom>

w«. :H. mesoCOHQ Hmcomhom

mo. mo. masocona o>avoomno

 

 

:N.- HH. moooz
:m.: *Hm.- unsosmmnm
:0. ma. onwpothCoo
HH. *oz. owedp nho>\csoz
0a.- 0a. 0a:
0H.- mH.: mCowmno>cfi ofi\mow
ma. No. mCOfiPmoso o:\wow
mo. *mo. mSOflPmmsv woe
Hm. mo.| mQOprosv 90:3
om.- mm.: wo>flpmnoQEH
mo.- mo.- wo>wponoaooa
D DAS nooomm nonpoz
mcfinomofio

 

.m

m manna

wCoHPmopflm oommpooofi> Ca moocopopps ocm Dds oaflno op nooomm nonpoe Mo mcowpmaohnoo Howpnmm

72

m0. v 9*

.mCoapmaopnoo omonp Eopw oo>oEon who mpoommo om¢.

 

 

 

 

mm.- 0N.-
0H. 00.-
Ho. 0m.-
m0. Na.-
00. ma.-
N0. NN.-
ma. mm.-
mH. N0.-
wN. #5:.
0N. 0N.
00. m0.
:0. *m:.-
m0. N0.-
MN. :0.
NH. Hm.-
mo. 00.-
z >

mace wcfi>oz

*mm.: $33.:
mH.- *mm.-
N0.- *Nm.-
mm. :m.:
m0. AN.-
mH.: *hm.:
3N.- ma.-
0H.- Ha.-
0N. 00.
00. 0N.-
0H. 3H.-
H0.- ma.-
0H.- ma.-
mH.- HH.
0H.- *0m.-
00.- 0H.-

2 >

mace Hmfloom

mH.- m0.-
0N.- H0.-
*Nm.- HN.
3.- mN.
mN.- 0H.
0N.- ma.
00. m0.-
NH.- ma.
0H.- 00.
0N.- no.
:0.- an.
Nfi.- 3N.
0H.- :0.
:N.- Nm.
mH.- 00.-
NN.- ma.
2 >

maop wcfl>oe::oz

 

8.- 8.-
Oo. HN.
2... Na.-
9: 2.-
Qool *mm..:
50.: :N.:
mo.: *md.:
n0.: mm.-
mm. Hm.
H0.: mm.
m«.: NH.:
m0. m0.:
mH.- m0.-
00. ma.-
20. 00.-
mm. 0H.-
2 >
wcflooom

 

0H.- H0.
m0.- N0.-
HN.- RN.-
sm.- *0m.-
2: mo...
00.: 00.
0m.- eN.-
Nfi.- Ma.-
No.- :m.
0H. *mm.
00. on.-
HH.- NH.-
HH.- NH.-
H0. m0.-
m0.- *N:.-
NH. 00.
z >
wCHnommHQ

mpoflchos npo>
whofieflooe :5oz
mano>

masoconm Honomnom
mCSoCOMQ o>wpoowno
moooz

mpCoemmnm
mcowpoonpcoo
owpon nuo>\c3oz
Dds

mCOflmno>cfi oo\wow
mGOfivmosv os\mow
mCoHPmosv woe
ozofipmosw 90:3
mo>apdhomEH

mo>apwhwaooo

 

nooomm nonpos

 

.m

m manna

macapNSPflm oommpooofl> Ca om: sz :30: van A>v nno> cafigo Op nooomw nonpoe no wCoHpmHlooo

73

mo. v Qt.

.mcoapmHoHHoo omonp Eon co>oEoH ohm mpoommo ow<m

 

 

Hm.: 0N.
Hm.: NM.
BN.I NO.I
mo.: ma.
0H.I HH.
NN.I 3H.
00.: 0o.
00.: mm.
0H. *mm.
ON.I OM.
mo.: 0H.
0H. 30.:
NO. 3N.
3m. OH.
*03.: 0H.:
0H.: ma.
0 ¢

mzop mcfi>o2
mcoapmspfim ooua> Cw om: ADV apaaowsmo 0cm A30 anmflafixso oafino op mCofiPmaop gooomm 90:90:

N

 

Hn.- Nm.- 0H.- N0. R0. 00.
0H.- 0m.- mN.- 0N.- 0N. 0N.-
0N.: mN.: HH.: 00. no. 0H.
N0. N0. 0H. 00. NH. Hm.
0N. 0H. 00. 0N. N0. 0H.
0H.: 0H.: 00.: no. 00. OH.
00.: 0N.: 0H.: mm.: 00. OH.-
30.: NH.: 00.: 0H.: 0N. om.
0N.: mm.: 3m.: 3a.: mm. mH.
N0. 0H.: 00.: Hm.: mH.: 0H.
0H.: NH.: mm. 0H. mm.: 0H.
H0. H0. 0H. Hm. mm. 00.
00.: 0H.: 00.: 00. 00.: OH.-
mo. NH. mH.- 00. *ms. 3N.-
mm.: 0H.: mm.: 00.: 00.: H0.
so. mH.- mo.- NH.- mm. 0H.
m»%p Hmflwom whopowcfl>om:cOz wcfiuoow

 

 

00.: NN.: mnoflmwpoe nno>
*03.: 0H.: mnowmaooe csoz
0N.: 0N.: mpHo>
00. 0H.: m:3o:onm Honomnom
HH.: 00. masoconm o>apoomno
mm.: no. mocoz
*mm.: HN.: mesoswmnm
00. 00. moofipomnpcoo
no. 00.: capmn Quo>\csoz
*33. 30. :92
0H.: 3H.: mCoamuo>Cw ofi\mow
N0.: 0H.: mCOfipwoSU ofi\mow
00. 3a.: mCowpmosv was
00.: 00.: mCoHpm030 Pwsa
00. NN. mm>H9mMmQEH
NH.I 3a.: wm>wvmhmdomo
0 < noomQW-Honvos
wcflpogmflo

 

OH manna

74
reliable pattern (minimum rz‘: .31, p “.1, 2 tailed test),
for mother to child speech across situations, though the
direction of the relationships were similar across situa-
tions. This correlational pattern containing few signifi—
cant relations is similar to the concurrent correlations of
mother to child speech at 27 months reported by Furrow et
al; mother imperative use positively related and mother in-
terjective (fragment) and verb use negatively related to
child MLU, nouns/utterance, and verbs/utterance. In the
present study, mother interjection (fragments) and verb use
negatively related to child speech, though mother imperative
use also negatively related to child speech. Situations
substantially altered observed relationships. Mother imper-
ative use negatively related to child speech in social and
moving toy situations; mother verb use negatively related to
child speech in the social toys situation.

Furrow et al predicted positive relationships for mother
speech to 18 month olds to child speech at 27 months for (a)
wh- questions, (b) yes/no questions, and (c) deletions (yes/
no questions with deleted frontal auxiliary verbs); negative
relations were predicted for mother speech variables of:

(a) declaratives, (b) MLU, and the frequency per utterance of
(c) sentence nodes, (d) pronouns, (e) verbs, (f) modifiers
(noun and verb), and (g) contractions. Mother to child
speech in the present study showed positive relations for

wh- questions and MLU, and negative relations for verbs,

noun modifiers, and verb modifiers across situations, with

75
significant relations appearing in the diapering, feeding,
social and moving toys situations. Thus, the pattern of re-
lationships from this cross-sectional study is similar to
Furrow et al's concurrent (mother speech to 27 months to 27
month old speech) relations. Mother grammatical speech cat-
egories were twice as likely to be related to child speech
than mother sentence categories (9.9% versus 5%, p < .05,or
15.3% versus 10.6%, p <.10).

The combined effect of mother speech and child cogni-
tive development to child speech is examined in a series of
exploratory multiple regression equations predicting 2 com-
plex measures of language development, MLU and causality
statements, and one measure of speech frequency, utterances,
produced in each of the 5 major videotaped situations. Age,
object permanence, means-end relations, and mother use of
yes/no questions, imperatives, and deletions (yes/no ques-
tions with deleted frontal auxiliary verbs) were used as pre-
dictors. Variables placed in the regression equations re-
quired attaining an F ratio at the 95th percentile of the F
distribution for 30 subjects (F = 4.17); once placed in the
equation a variable was dropped if its F ratio (predictive
value) fell below the 90th percentile of the F distribution
(F = 2.88, n = 30). The variable age was not subject to
these statistical requirements; age automatically entered
the equations as the first predictor to eliminate age var-
iance from all subsequent predictors. Table 11 lists the

results from the multiple regression equations.

76
Table 11
Multiple Rs (MR) and Beta weights predicting children's
speech in videotaped situations with age, object permanence,
means-end relations, mother yes/no question and imperative

(IMP) use as predictors.

 

 

 

 

Situation Child speech MR Predictor E213
Diapering MLU .46** age .46
Utterances .39* age .39
Causality .53** age .53
Feeding MLU .28 age .28
Utterances .65** age .65
Causality .50** age .50
Non—moving toys MLU .55** age .55
Utterances .37* age .37
Causality .42* age .42
Social toys MLU .52** age .32
IMP -.35
Utterances .52** age .52
Causality .38* age .38
Moving toys MLU .52** age .19
IMP -.44
Utterances .59** age .41
IMP -.35
Causality .64” IMP - . 1:2
age .40

 

*p < .05
**p < .01

77

Age predicts MLU in the diapering, non-moving toys, and
social toys situations; imperatives are significant predict-
ors in the social and moving toys situations. Examination
of the Beta weights show the impact of the imperatives as
reducing MLU in these situations.

Age also significantly predicts utterances in all sit-
uations. Imperatives significantly predicted utterances in
the moving toys situation; Beta weights show the impact as
negative, though this reduction may be related to the gen-
eral reduction of utterances reported earlier for this sit-
uation.

In sum, hypothesis 4 is not supported. There is no ap-
parent combined relationship of cognitive development and
mother speech that is positively related to child speech;
mother imperative use is related to decreases in MLU, utter-
ances, and causality statements in the moving toys situation,
and to MLU in the social toys situation. Parial correlations
of mother speech to child speech have relations that agree
with concurrent and predictive correlations reported by
Furrow et al. Implications and interpretations of these re-

lationships are made in the discussion.

DISCUSSION

The present study explicitly examines 4 variables
thought to contribute to child language and speech develop-
ment: (a) situational effects, (b) object permanence,

(c) means-end relations, and (d) mother speech, and one
implicit variable, age. Age related to most aspects of

child speech, language, and cognitive development. Many
first order correlations between variables were statistically
reliable when age factors were not controlled; few relations
were reliable once age factors were controlled. Newport et
a1 (1977) reported similar findings in which most relations
were substantial until age and initial child speech factors
were controlled in their longitudinal study; since the pre-
sent study was cross-sectional in nature only age factors
were controlled.

Hypothesis one

This hypothesis measures child speech activity across
situations in a series of analyses of variance.

The present situational analysis is consistent with the
findings of Benedict and Truhon (note 1); child utterances
decreased in the moving toys situation which is comparable
to the decrease in utterances in the complex shape toy situ-

ation of the Benedict and Truhon study. MLU did not differ

78

79
across situations in the present study, replicating Benedict
and Truhon's finding of no MLU differences across book read—
ing, free play, and directed play (complex shape toy) situ-
ations. More engrossing situations are associated with de-
creases in child utterances, probably due to the children
attending or acting upon the complex display.

It appears that children presented with an interesting
phenomenon will concentrate on that phenomenon, perhaps be-
cause of its novelty or in an attempt to understand its
principles, and in doing so reduce their speech. Utterance
reduction may be an attempt by the infant to functionally
increase its ability to gather information about an inter-
esting event. When the children do talk, it appears the sit—
uation does not affect the general complexity of speech (MLU)
though components of speech, expecially verbs/utterance do
change across situations.

Infant interaction with each toy was short, lasting be-
tween one—and-one-half and three minutes for each situation;
short periods reduce potential utterances which may affect
child speech characteristics. In post taping interviews,
all mothers reported engaging their infants in a manner sim-
ilar to the filmed sequences. Most situations were completed
before reaching the full amount of time allowed. The short
time period for each situation appears typical of mother to
infant play interactions, indicating infant speech was appro-

priately measured.

8O

Hypothesis two

Relations of object permanence to child language and
speech are examined in hypothesis 2.

All significant relations of child vocabulary and ob-
ject permanence occurred in receptive categories; these
categories were: (a) features, (b) recurrence, and (e) per-
son to object locative relations. Relations of receptive
vocabulary categories to object permanence were higher than
corresponding production vocabulary categories in 10 or 12
categories (exceptions were: (a) total words, gs = .15 ver—
sus .26, and amounts, pg 2 .15 versus .15). These partial
correlations, though most were not significant, are consis-
tent with the view that language comprehension is "more
directly in line" to basic concepts or cognitions under-
lying a language category (Cromer, 1979).

Cromer (1979) hypothesizes cognition is the product of
two interacting factors: (a) structural factors consisting
of components such as short term memory, and (b) operation
factors consisting of sensorimotor and cognitive abilities
such as object permanence and reversibility. Linguistic
developments interact with cognitive developments when a con-
cept first forms; further developments procede as a function
of linguistic properties. Cromer supports his hypothesis by
describing Bellugi's (1971) findings on self reference.
Bellugi found children: (a) initially use their own names
for self reference, (b) gradually acquire the pronouns "I"

and "me" but incorrectly place these pronouns in their

81

utterances, and (c) correctly use the pronouns within their
utterances. Increased self reference sophistication as in-
dicated in Bellugi's sequences does not show a change in the
children's self awareness, but only how self is expressed to
others. Cromer's hypothesis would predict higher object per-
manence relations to receptive language compared to product-
ive language.

Cromer's hypothesis requires some delineation of the
unique language development factors which would cause high-
er relations for receptive language. Menyuk and Menn (1977)
hypothesize language development consistsof 2 processes,
comprehension and speaking. Menyuk and Menn view language
development as a phenomenon in its own right, though cogni-
tive and social factors that impact on language are included.
Their model includes: (a) social perception, described as
learning words within situations and knowing when to pro-
duce these words, (b) memory, described in terms of recal-
ling the appropriate sound pattern constituting a word for
a situation, and (c) cognition, described as using words to
reach goals within novel situations. Their full model is
outlined below.

A. Comprehension processes
1. Readiness in processing speech
a. distinction between speech and nonspeech
b. discrimination between speech sound cate—
gories and prosodic patterns

c. production of consonant—vowel sequences
marked prosodically

82
2. Analysis of meaningful unit(s)

a. phonological sequences and situation

b. phonological analysis and situational
analysis

0. analysis of syllabic features of
phonological sequences

d. analysis of distinctive feature
content of sequences

B. Speaking processes

to learn to produce a variety of vocal sounds

to learn to produce vocal sound patterns so

that they more or less match sounds which are

heard

5. to learn to remember certain sound patterns
well enough to produce them without just
having heard them

6. to learn to produce specific sound patterns
in situations where they have been produced by
others or by oneself in the past

7. to become able to produce a word in a novel

setting as a means to an end

cm»

The Menyuk and Menn model is a descriptive model of
language and speech development, which does not specify in-
itial concept formation as described by Cromer; the model
does emphasize additional processing stages needed for lan-
guage production. Presumably, children comprehending lan-
guage may need to complete processing of the first 2 steps
plus some minimal level of object permanence development.
Language production requires completing new steps (3 to 7).
Complex or abstract speech requires increased knowledge in
means-end relations (step 7). and social perception requires
knowing when to speak (step 6). Practice effects are out-
lined in steps 3, 4 and 5 for all types of production (chil-
dren in the present study did not need to produce fully
formed adult-like words for vocabulary inclusion). Tenta-

tively, productive vocabulary categories appear more

83
"distanced" than corresponding receptive vocabulary categor—
ies to object permanence development; processes similar to
those described by Menyuk and Menn may explain this dis-
tancing.

The 3 receptive language categories related to object
permanence all contain words that modify object character-
istics and locations; knowledge implied by knowing these
words should relate to object permanence performance. The
first category, features, consists ofwcrds that modify ob—
jects, such as big, little, broken, or dirty. Children

knowing these terms should know more about distinctive parts
of objects which may assist in recall of that object, or in
object permanence, an awareness that a particular object was
hidden which can enhance continued search for that object.
The second category, recurrence, is made up of words that
indicate objects and actions that disappear and return, such
as more and again. Children knowing these words have aware-
ness of hiding situations such as those used in object per—
manence. The final category, person to object locative re-
lations, includes words marking locations relative to the
child, such as up, down, front, back, or side. Children
knowing these words are aware of spatial locations in refer-
ence to themselves, knowledge that enhances their object
permanence performance. Increased "objective" locative
knowledge, as indicated by knowing words marking locations
from a vantage point away from the child, is not related to
object permanence performance; "objective" locative words

may relate to higher order cognitions such as reversibility,

84
but for object permanence a knowledge of self to locations
appears necessary.

Receptive vocabulary categories, which did not relate to
object permanence performance, but are associated with ob-
jects, were nouns, proper nouns, colors, amounts, possessives,
and nonexistence. Ramsey et a1 (1978) report first object
words are acquired with acquisition of stage 5 of object
permanence; relations of nouns, which tend to code object
words, may have been attenuated since most children in this
study were assessed at a stage 6 level, well past the time of
initial concept acquisition. Latter acquisition of nouns may
develop as a function of linguistic factors (Cromer, 1979;
Menyuk & Menn, 1977). or factors not related to object per-
manence development. Object related categories not appearing
to tape knowledge required in object permanence performance
are: (a) amounts, such as all and some, which focus on groups,
(b) nonexistence, which logically relates to object disappear-
ence, but not knowledge that assists in the recovery of ob-
jects, and (c) possessives, which denote ownership. Colors,
like features, should relate to object permanence as a way to
recall characteristics that make hidden objects distinct; a
focus on colors may not be efficient for this task.

Object permanence did not relate in taped situations to
speaking words coding: (a) recurrence, (b) nonexistence, and
(c) causality statements; a result consistent with the find-
ing that object permanence is not related to production voc-

abulary categories. Speaking words coding nonexistence and

85

recurrence have been reported as occurring after attaining

the highest levels in object permanence (Corrigan, 1978), and
stage 6a by McCune-Nicholich (1981); these reports do not use
correlations but rather mark the occurrence of the speech af-
ter cognitive development. Most children in the present

study performed at object permanence stage 6b (24/30); this
large number of children may have attenuated relations be-
cause of a time lag between the minimum conceptual level need-
ed for acquisition and the resulting "drift" as both areas de-
veloped. The present study supports McCune-Nicholich's find-
ing that words coding nonexistence and recurrence occur after
achievement of object permanence stage 6a. Diary records

show stage 6a children producing and comprehending words cod-
ing recurrence (pg = 1, 2), and nonexistence (pg = 1, 3);
though only stage 6b children produced these words in taped
situations.

Object permanence positively related to nouns/utterance
in free play (g = .38, p«<.05). This relationship may be
linked to the large number of toys in free play; 11 children
frequently held up toys to show to mothers and experimenters,
then named the toys. These children named toys in free play
(typically 2 to 5 toys) up to 10 times apiece.

In sum, object permanence is related to a small set of
receptive vocabulary categories; a model of "distance" from an
initial concept acquisition to receptive and productive vocab-

ulary build-up appear to best explain this pattern.

86

Hypothesis three

Hypothesis 3 examines relations of means-end relations
to child speech and language development. Means-end re-
lations negatively related to child verbs/utterance in most
taped situations; marginally postitive relations to causal-
ity statements were found in 2 situations. Lower levels of
means-end relations were related to increased verb use,
while higher levels marginally and positively related to
increased causality statements.

Correlations of means-end relations scores from the
first administration to the second administration was .93;
three children changed their scores, 2 increased and 1 de-
creased though changes were small, comprising of no more
than 2 scale steps.

The 6 subjects performing within stage 6 on this mea-
sure were inconsistent since 2 children succeeding in the
ring situation did not solve the necklace situation, a
lower scale step. Examination of the subjects' stage 5
behaviors show consistency across steps; 2 of 30 subjects
had out of sequence scores and both subjects did not pull
a string to get a vertically displaced toy, but used sticks
to extend reach. Means-end relations appears to accurately
and consistently measure reaching activity using implements,
and less consistently measures the planning and execution
of acts to obtain unseen goals. This inconsistency in stage
6 steps is important as stage 6 processes coordinate planning

and execution of actions in the "physical world"; processes

87
hypothesized as underlying the ability to make and talk
about causal inferences and request making to independent
causal agents (Bates, 1976; Snyder, 1978).

The standardized means-end relations scale (Uzgiris &
Hunt, 1975) was used in the present study; only 2 steps in
this measure assess stage 6 activity. The inconsistencies
of stage 6 performance may have attenuated relationships
of means-end relations to causality statements in the
present study; increasing the number of stage 6 steps in
this measure may help clarify any existing relationship.

Means-end relations marginally and positively related
to causality statements in 2 situations, diapering and
social toys. The similarity of activities in these 2 situ-
ations may in part explain these relationships; all 18 and
22 month olds removed some article of clothing from a doll,
and 7 children attempted to diaper a doll (5 22-month-olds
and 2 18-month-olds). Diapering may present a familar situ—
ation to the children to expect certain activities, facili-
tating talk about these activities. Diapering activities
include taking off clothing, powdering, locating new clothes,
and putting on the new clothing. Speaking a number of caus-
ality statements within a situation may require facilitating
situational support as well as cognitive activity.

Causality statements in feeding may have been reduced
since mothers often offered children more food during feed-
ing, reducing the need to request something from a separate

causal agent. Mother use of all forms of yes/no questions

88

were highest in feeding (40.7%) compared to all other situa-
tions (diapering = 26.8%; non-moving toys = 32.9%; social
toys = 28.5%; moving toys = 20.8%). Most (84.3%) of all
yes/no questions in feeding were related to food requests.

Causality relations in the non-moving toys situation
may have been attenuated since the toys did not possess
moving or removable parts; reducing the opportunities for
action requests and questions. The moving toys situation
contained toys capable of independent action and presumably
presented many opportunities for requests and causal state-
ments; means—end relations was not related to causality
statements in this situation. The situation may have been
too difficult for many of the children who may have been
engrossed with the toys reducing complex elements of their
speech. This interpretation is elaborated upon following
a discussion of verbs produced per utterance.

The negative relations of means-end relations to verbs/
utterance in diapering, feeding and moving toys situations
are linked to increased verb use by males in diapering and
moving toys situations, and to increased verb use by females
in feeding. Since only 6 subjects achieved stage 6 in means-
end relations, the increased verbs/utterance can be attrib-
uted to stage 5 children. Increased verb use by stage 5
children maybe related to inefficient and protracted goal
reaching and tool using activities in diapering, feeding
and moving toys situations. In the moving toys situation,

the verbs ”go" and "push" (or their equivalent) were spoken

89
most often during play with the car and top; these verbs
mark the most salient action of the toy in their respective
situation. Verb use in the moving toys situation make a
descriptive account of events in that situation; an ex-
planatory account of events in the moving toys situation is
linked to the production of causality statements which for
this situation may have been attenuated since children were
still forming their explanatory hypotheses.

In sum, means-end relations is related to many child
speech behaviors, and is not related to receptive language.
This is the opposite pattern found for object permanence
relations to language. Bates (1976) and Snyder (1978)
hypothesize means-end relations are linked to tool use and
the developmental use of intermediate goals to attain an
end goal in language and object manipulation for stage 5;
higher level inference making and causally oriented request
making are hypothesized for stage 6. The findings of the
present study support these views, especially when verbs
are considered as descriptive markers of events. Situations
altered the relations of verbs and causal statements; differ-
ences in the type of interaction found across situations are
hypothesized as causing these differences.

Hypothesis four

Hypothesis 4 compared the relations of mother speech

and child cognitive development to child speech development.

Mother speech data in videotaped situations consist of

60 mother utterances in the combined toy situations, and 30

9O

utterances in diapering and feeding; Newport et al (1977)
examined more than 100 utterances in a free play situation.
The lower number of utterances were a consequence of the
taping procedures as described in the methodology. Fewer
utterances may increase sampling error since less data are
analyzed. However, most situations were stopped by mothers
before the maximum time allowed. Interview indicate most
mother-child play interactions are described as being of
short duration. Klein (1980) reports the typical mother-
child interaction lasts a few minutes, not the 10 or more
minute interaction often used in assessing mother-child
speech relations. In the present study, only 6 mothers
engaged in prolonged free play with their children. General-
ly, mothers did not interact with their children except for
smiling and encouraging them to play until the children
requested assistance or when mothers believed help was need-
ed. Most mothers (p = 27) stated in interviews that they
would usually engage in similar interactions with their chil-
dren, a general monitoring of their children's activity with
focused intervention when needed. Intervention consisted of
accomplishing a specific purpose such as straightening
clothes, comforting, or showing how a toy worked.

Mother to child speech relations across situations show
a pattern similar to concurrent (for mother use of fragments
and verbs) and predictive (for mother use of nodes, pronouns,

verbs, modifiers) relations presented by Furrow et al (1979).

Most mother sentence categories were not related to child

91
speech, except for imperatives.

Increased use of imperatives relates to decreases in
child MLU and verbs/utterance, especially in diapering,
moving toys, and social toys situations. The negative
relations suggest mother imperative use caused children to
attend to their mothers, or carry out the action requested
in the imperative effectively diverting the children's
attention away from speaking.

Mother grammatical speech categories tend to show nega-
tive relations for inherently complex and abstract categories
(fragments, nodes, pronouns, modifiers, verbs), and positive
relations for categories that can be fine tuned (MLU, noun/
verb ratio) to levels of child speech; these relations change
across situations but not dramatically, for example, positive
correlations in one situation for a set of variables tend to
be positive in another situation.

The pattern of cross—sectional relations of mother to
child speech is important for making longitudinal inferences.
The present results show mother speech complexity, abstract—
ness, and intent relate to child speech; the size of these
relations qualify longitudinal relations. Furrow et al's
relations of mother speech to 18 month olds correlated to
child speech at 27 months, were not the same as the mother
to child speech relations at 27 months. However, Furrow
et al did not present cross-sectional mother to child re-
lations at 18 months, nor were relations of child speech at

18 months to child speech at 27 months presented. Furrow

92
et al studied children who were equivalent in language skill
(MLU about 1.0); differences in their sample's comprehension
skill were noted but not described. It is possible that
mother speech is, at least partially, related to aspects
of child behavior not involved in actual child speech. Moth-
er speech maybe related to the child's comprehension skill,
cognitive development or vocabulary knowledge; skills and
developments in these non—speech areas may provide the basis
for latter speech development. In this view, mother speech
occurs as a function of situational characteristics and
feedback from children's non-speech and speech behaviors,
and that long term development of speech occurs as a function
of conceptual and language comprehension developments.

The impact of child non-speech relations to mother
speech can be examined in mother imperative use. Imperative
use negatively related to child MLU in social and moving toys
situations (pg = -.37 & -.45, p < .05); imperative use pos-
itively related to child relational comprehension ability
(as assessed in the third and fourth home visits) to the
same situations (pg = .54 & .40, p < .05), the remaining
relations in the other situations were not significant (pg
range from -.08 to .16, p < .10). The social and moving
toys situations had relatively complex toys which could be
actively manipulated and explored. The procedures for re-
lational comprehension assessment emphasize complex impera-
tive—like commands. Mothers in the social and moving toys

situations may have used imperatives which were successful

93
in previous encounters with complex toys, increasing their
children's understanding or activity in a situation but
decreasing their speech.
General summary and conclusions

The present study confirms previous findings that gen-
eral speech complexity as measured by MLU does not change
for young children across a variety of situations. However,
specific speech forms do differ across situations (verbs/
utterance); these differences may be linked to cognitive
development (means-end relations) and attempts to understand
or describe ongoing events. Causality statements are margin-
ally related to cognitive development (means-end relations);
situational characteristics and task familiarity may provide
opportunities to demonstrate this language behavior.

Object permanence is related to some object oriented
vocabulary categories, but only in comprehension. These
object categories appear to reflect an awareness of concepts
that are useful for success in tests of object permanence.

Many forms of mother speech relate to child speech. The
general pattern is that inherently complex speech negatively
relates to child speech; fine tuned speech positively relates

to child speech. Also, some mother speech relate to child

relational comprehension.

REFERENCE N OTE

REFERENCE NOTE

1. Benedict, H., & Truhon, S., Context effects on early
infant language, unpublished manuscript, 1981.

9L1

LIST OF REFERENCES

LIST OF REFERENCES

Bakeman, R., & Brown, J., Early interaction: consequences
for social and mental develoBment at three years. Child
Development, 1980, 5;, 437-4 7.

Bates, 2., Language in context. New York: Academic Press,
197 .

Bayley, N. Manual for the Bayley scales of infant develpp—
ment. The Psychological Corporation, New York, 1969.

Bayley, B., & Schaefer, 5., Correlations of maternal and
child behaviors with the development of mental abilities:
data from the Berkeley growth study. Monogpaphs of the
Society for Research in Cpild Development. 1964, pg,
serial 97.

Benedict, H., Language comprehension in 10- to 16-month old
infants, unpublished doctoral dissertation, Yale Univer-
sity, 1976.

Benedict, H., Early lexical development: comprehension and
production, Journal of Child Language, 6, 183-200, 1979.

Bellugi, U., Simplification in children's language, in
Huxley, R., & Ingram, E., (Eds), Language acquisition:
models and methodg, Academic Press, New York, 1971, 95—
119.

Bloom, L., One word at a time. The Hague: Mouton, 1973.

Bowerman, M., Structural relationships in children's utter-
ances; syntactic or semantic? in T.E. Moore (Ed.)

Cognitive development and the acquisition of langugge.

New York: Academic Press, 1973.

Brown, R., A first language: The early stages. Cambridge,
Mass: Harvard University Press, 1973.

Bruner, J.S., The ontogenesis of speech acts. Journal of
Child Languag_. 1975, g, 1-19.

Corrigan, R., Language development as related to stage 6
object permanence development, Journal of Child Languag_.
1978. 5. 173-189-

 

95

96

Cromer, R., The strengths of the weak form of the cognition
hypothesis for language acquisition, in Lee, V., (Ed.),
Language Development. John Wiley & Sons, New York, 1979,
102-130.

Elardo, R., Bradley, R., & Caldwell, B.M., The relation of
infants' home environments to mental test performance
from six to thirty-six months: a longitudinal analysis,
Child Development, 1975, 4Q, 71-76.

Furrow, D., Nelson, K., & Benedict, H., Mothers' speech to
children and syntactic development. Journal of Child

Language, 1979, g, 423-442.

Golinkoff, B.M., & Ames, G.J., A comparison of fathers' and
mothers' speech with their young children. Child Devel—
Cement. 1979. 29, 28-320

Greenfield, P. M., & Smith, J.H., The structure of communi-
ggtion in early langugge develppment. New York: Academic
Press, 1976.

Halliday, M.A.K., Learning how to meap:explorations in the
development ofglgnguage. London: Edward Arnold. 1975.

Hood, L., & Bloom, L., What, when, and how about why: A
longitudinal study of early expressions of causality.
Monographs of the Society for Research in Child Develpp-
ment, 1979, 44, 6, serial 181.

Ingram, D., Sensorimotor intelligence and language develop-
ment, in Lock, A., (Ed), Act10p1 Gesture and Symbols:
The emergence of languag_. New York: Academic Press,

1976.

Jackson, E., Campos, J., & Fischer, K., The question of
decalage between object permanence and person perman—
ence. Developmental Psychology, 1978, $4, 1, 1-10.

Klein, D., unpublished doctoral dissertation, Michigan State
University, 1980.

Lieven, E. Conversations between mothers and young children:
Individual differences and their implications for the
study of language learning. In N. Waterson & C. Snow
(Eds.), The development of communicationggocial ape
pgggpatic factors in language acquisition. London: Wiley,
1 .

Macnamara, J., Cognitive basis of language learning in infants,
Psychological Review, 1972, 79, 1—13.

McCall, R., Appelbaum, M., & Hogarty, R., Developmental

97

changes in mental performance. Montraphs of the Society
for Research in Child Development, 1973, 32, 3, serial
150.

McCune-Nicolich, L., The cognitive bases of relational words
in the :ingle word period, Journal of Child Language,
Q9 15-3 '

Nelson, K., Concept, word, and sentence: interrelations in
acquistion and development. Psychological Review, 8;,
267-285, 1974.

Nelson, K.E., Aspects of language acquisition and use from
2 to age 20, Journal of the American Academy of Child
Psychiatpy, 1978, 583-607.

Newport, E., Gleitman, L., & Gleitman, H., Mother I'd rather
do it myself: some effects and non-effects of maternal
speech style, in C. Snow & C. Fuerguson (Eds.) Talking
to children: language input and acquisition. Cambridge,

1

C.U.P. 1977.

Ninio, A., & Bruner, J., The achievement and antecedents of
labeling, Journal of Child Language, 1978, 5, 1-15.

Paraskevopoulow, J., & Hunt, J., Object construction and
imitation under differing conditions of rearing,
Journal of Genetic Psychology, 1971, 112, 301-321.

Piaget, J., The construction of realipy in the child. New
York: Basic, 1954.

Piaget, J., The origins of intelligence in children. New
York: Norton, 1963.

Piaget, J., Six psychological_gtudies. New York: Random
House, 1967.

Ramsey, R.E., Campos, J.J., & Fenson, L., Object word vocab-
ulary and stage 6 object permanence in infants. Paper
presented at the southeastern conference on Human devel-
opment, Atlanta, Georgia, April, 1978.

Ratner, W., & Bruner, J., Games, social exchanges and the
acquisition of language, Journal of Child Language,

1978. 5. 391-401-

Rodgon, M., Single word usage, cognitive development and the
beginnings of combinatorial speech. New York: Cambridge
University Press, 1976:

Schlesinger, I.M., Production of utterances and language
acquisition, in D. Slobin (Ed.), The ontogenesis of

98

gramma . New York: Academic Press, 1971.

Shatz, M., On the development of communicative understandings:
an early strategy for interpreting and responding,
Cognitive Psychology, 1978, pp, 271-301.

Shatz, M., Children's comprehension of their mothers'
questifig-directives, Journal of Child Language, 1978,
it 39" '

Sinclair, H., Sensorimotor action patterns as a condition
for the acquisition of syntax, in R. Huxley & E. Ingram
(Eds.), Language acquisition: models and methods.

London: Academic Press, 1971.

Snow, C., The development of conversation between mothers
and babies, Journal of Child Language, 3, 1977, 1-22.

Snyder, L., Communicative and cognitive abilities and
disabilities in the sensorimotor period, Merrill-

Palmer Quarterly, 1978, gfly 3, 161-181.
Ungerer, J., Selazo, P., Kearsley, R., & O'Leary, K., Devel-
opmental changes in the representation of objects in

symbolic play from 18 to 3 months of age, Child Devel-
Uzgiris, I., Patterns of cognitive development in infancy,
Merrill-Palmer Quarterly, 1973, $2, 181-204.

Uzgiris, I., & Hunt, J., Assessment in infancy: orginal
scales of psyehological development. Urbana: The
University of Illinois Press, 1975.

Zachery, W , Ordinality and interdependence of representations
and language development in infancy, Child Develppment,

1978, 4 , 81-687.

APPENDICES

APPENDIX A

APPENDIX A

LANGUAGE COMPREHENSION ASSESSMENT

The language comprehension assessment consist of 21

stock sentences written in either a modified imperative

or question form. The sentences contrast a variety of verb

and object names; the object names are left blank in the

sentences so that particular object object names known to

the children can be inserted into the sentences for con-

trasts.

low.

H
o0ooxi0xntum-a

HHH
\ONH

14.
16.
17.
18.
19.
20.
21.

H
U1

The
tences;

tences.

Each of the 21 stock sentences are reproduced be-

. Where's the 0-1 and the 0-3?

Get Mommy's shoe.

Give R-2 the broken O-C.
Put the 0-4 in the L-1.
Where's (child's name) C-1?

. Get mommy the shoe.

. Throw me your ball.

. Get the 0-F2 from mommy.
. Where's mommy's C-2?

. Get the little O-C.

. Get the OF1 for mommy.

. Give the big O-B to R-2.

Get (examiner's name) the book.
Give the 0-2 and 0-3 to mommy.
Show me mommy's C-2.

Give the 0-1 to R-2.

Get (examiner's name)'s book.
Put the 0-2 in the L-1.

Get the 0-3 and 0-4.

Give me the O-1 and 0-2.

Put the pillow on the baby.

symbols represent words to be inserted into the sen-
these symbols can also mark the intent of most sen-

The "0" words with numbers represent toy contrasts

(0-1, 0-2, 0-3, 0-4); "0" words with letters represent size

99

100
or broken-object contrasts (O-B = broken object). The
"0" words with letters and numbers represent contrasts of
"for" and "from”. The "R" words contrast recipients and
the "L-l" word represents a location. No objects or people
are repeated across different symbols; for example, 0-1 may
be a small toy car for sentences 1, 16 and 20 but that car
cannot be used for any other sentence.
Administratipp,procedures

Object names needed for the stock sentences were obtained
during the first and second home visits from parent interviews
and from the diaries and checklists. The third or fourth
home visits were scheduled at a time when a recipient (for
sentences 3, 12 & 16) could be present.

On the testing day, all toys except those used in the
contrasts were removed from the room in which the contrasts
were to occur. The contrast toys were usually placed in a
pile on the floor of that room. The sentences were written
in bold ink on index cards (one sentence to a card). Mothers
were given the cards one at a time and told to read the
sentence in a natural manner as possible when prepared. Each
trial lasted for 60 seconds; mothers were instructed to look
at the floor after each reading and not to repeat the sentence
until 60 seconds had elapsed. Each sentence was read up to
five times apiece. Fewer repetitions were read on trials
where disinterest or successful responding to the sentences

had occurred.

101

Scoringpprocedures

For each sentence and its' repetitions, the following
responses were recorded by the 2 experimenters.

1. Objects looked at by the child.

2. Objects pointed to by the child.

3. Objects touched by the child.

4. Objects picked up by the child.

5. Objects given to various people, or objects

placed in various locations by the child.

6. Words spoken by the child.

Discrepancies were resolved by discussion after each
trial. In initially correct responses, the sentence was
repeated a second time latter that session. The best
response to a sentence was taken. Contrasts were latter
compared across sentences. For example, responses to toy
cars were noted as was differentiation to small and large
cars; similarly differentiation for broken objects, such

as broken and intact balls, was noted.

APPENDIX B

APPENDIX B

OBJECT PERMANENCE ASSESSMENT

The object permanence assessment combines procedures

derived from the Uzgiris and Hunt (1975) scaling by Corrigan

(1978) and Jackson et al (1978). There are 4 main levels

of task difficulty:

(a) visible hiding, (b) visible dis-

placement, (c) invisible hiding, and (d) invisible displace-

ment. Each levels contains 2 or 3 locations for object

hiding or displacement. Object permanence steps are sum-

marized below.

Level I: visible hiding

1. Hide
2. Hide
3. Hide

then
4. Hide
5. Hide

object at A, infant finds object.
object at B, infant searches at A.
object at B, infant searches at A,
B.

object at B, infant goes to B first.

object at A or B alternatively,

infant searches initially at correct location.

6. Hide
fant

object at A, B or C alternatively, in-

searches initially at correct location.

Level II: visible displacement

7. Hide
hand
8. Hide
hand

object under A, then place object by
under B, infant searches only at A.
object under A, then place object by

under B, infant goes to A then B.

102

103

9. Hide object under A, then place object under
B; infant searches directly at B.
10. Hide object under A, then place object under
B; infant searches directly at B when direc-
tion of displacement occurs in either dir-
ection.
11. Hide object under A, then displace object to
B and then C; infant searches at A or B.
12. Hide object under A, then displace object to
B, then C by hand, infant searches at A or
B before going to C.
13. Hide object under A, then displace object to
B and C by hand, infant searches directly at
C.
14. Hide object under A, then displace object to
B, then C by hand; infant searches directly
at C when displacement occurs in either dir-
ection.
Level III: invisible hiding
These steps are the same as those described for
visible hiding except the object is first placed inside
a small container in front of the attentive infant be-
fore hiding; after hiding the object at the terminal
location for a trial, the empty container is shown to
the infant.
15. Hide container at A, infant finds object at A.

16. Hide container at B, infant searches at A.

17.

18.

19.

20.

Level IV:

104

Hide container at B, infant searches at A
before finding object at B.

Hide container at B, infant finds object

by first searching at B.

Hide container at A or B alternatively,

infant searches initially at the correct
location.

Hide container at A, B or C alternatively,
infant searches initially at correct location.

invisible displacements

These steps are similar to those described for

visible displacements, except the object is first

placed inside a small container in front of the

attentive

infant before hiding; after hiding the object

at the terminal location for a trial, the empty con-

tainer is

21.

22.

23.

24.

shown to the infant.

Displace container from A to B and hide
object at B; infant searches only at A.
Displace container from A to B and hide
object at B; infant initially searches at
A before finding object at B.

Displace container from A to B and hide
object at B; infant first searches at B.
Displace container from A to B and hide
object at B; infant first searches at B
from 1 direction only, searches from A to B

from the other direction.

105

25. Displace container from A to B and hide
object at B; infant can search initially
at B from either direction of displacement.

26. Displace container from A to B then C, hide
object under C; infant searches at A or B.

27. Displace container from A to B then C, hide
object under C; infant searches at A or B
before finding object at C.

28. Displace container from A to B then C, hide
object under C; infant searches at C initially.

29. Displace container from A to B then C, hide
object under C; infant searches at C initially
from 1 direction only, searches from A or B
then C in the other direction.

30. Displace container from A to B then C, hide
object under C; infant searches at C initially
from either direction of displacement.

31. Displace container from A to B then C, hide
object under A; infant searches at C initially
before searching at B then A.

These object permanence steps were repeated across 2
sessions. Children were counted as attaining a step when

passing that steps 3 times in 1 session.

APPENDIX C

APPENDIX C

Table A
Children 14, 18 and 22 months old passing (p) and

not passing (n) means-end relations scale steps

 

 

 

Extend Box and Red Filled
A_3 §e§_ Ha Vb Reach C Necklace Cage Stick Board Ring
14 f p n n n n n n n
14 f p n n n n p n n
14 f n n n n n n n n
14 f p n n n n n n n
14 f p n n n n n n n
14 m n n n n n n n n
14 m p p n n n n n n
14 m p n n n n n n n
14 m p n n n n n n n
14 m p p n n n n n n
18 f p n n n n n n n
18 f p p n n n p n n
18 f p n n n n n n n
18 f p p n n n n n n
18 f p p n n n n p n

 

aThis step is the horizontal string pull.

bThis step is the vertical string pull.

0This step is extending reach using a stick.

106

107

Table A (cont'd)
Children 14, 18 and 22 months old passing (p) and

not passing (n) means-end relations scale steps

 

 

 

 

 

 

 

 

 

 

 

a b ExtendC Box and Red Filled
Age Sgp, H 1 Reach Necklace Cage St1ck Boapg R1ng
18 m n n n n n n n n
18 m n n n n n n n n
18 m p n p n n n n n
18 m p n n n n n n n
18 m n n n n n n n n
22 f p p p p n n n P
22 f P P P P n P P n
22 f p p n n n P n P
22 f p p p p n p n p
22 f p p p p p n n n
22 m p p p n n P P P
22 m p n n n n p n n
22 m p p p n n p n n
22 m p n n n n p n n
22 m p n n n n P P n
totals 53% 40% 23% 13% 3% 33% 13% 13%

 

aThis step is the horizontal string pull.
bThis step is the vertical string pull.

0This step is extending reach using a stick.