RECALL AND CQMPREHEHSEQN AS A
FUNCTEQR QF MQDE Q5 FFESEN‘FATEQN
AND W [\ELQA «AUG “$14k! CQMFLEXEW

IN CQMRECTED iiE‘CGU-Ffifi

Tim‘sts Few Hm Dogma of pit. D.
WCH’LGAN STATE UNIVERSITY
Frederick Moses Okatcha
1968

LIBRARY 7

l-HESIS Michigan State
University

 

This is to certify that the

thesis entitled

Recall and Comprehension as a Fanction'of
Mode of Presentation and Transformational
Complexity in Connected Discourse

presented by

Frederick Moses Okatcha

has been accepted towards fulfillment
of the requirements for

M degree inEanaLiDnal Psychology

1
Major proztssdr V
Date M759

0-169

RECALL AND COMPREHENSION AS A.FUNCTION OF
MODE OF PRESENTATION AND TRANSFORMATIONAL
COMPLEXITY IN CONNECTED DISCOUBSE

By

Frederick Moses Okatcha

A.THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements

for the degree of

DOCTOR OF PHILOSOPHY

Department of Educational Psychology
1968

ABSTRACT

RECALL AND COMPREHENSION AS A FUNCTION OF
MODE OF PRESENTATION AND TRANSFORMATIONAL
COMPLEXITY IN CONNECTED DISCOURSE

By
Frederick Moses Okatcha

The main objective of this study was to investigate
variables which possibly influence recall and comprehension
of prose material. More specifically, the effects of number
of presentations and rate of presentation on recall and
comprehension of low difficulty and high difficulty prose
passages were examined.

Massed, distributed, and slow one-trial modes of presenta-
tion were employed in the study.

A total of one hundred and twenty gs were assigned to
6 experimental conditions.

The results indicated that number of presentations was
not a significant factor facilitating learning of prose
material. Significantly poorer recall and comprehension
was associated with prose material containing complex gramma-
tical constructions. However. there was a significant

interaction between mode of presentation and level of

11

Frederick Moses Okatcha

grammatical complexity. The slow one-trial mode greatly
facilitated the recall and comprehension of grammatically
complex prose material.

The results were interpreted as supporting a transforma-
tional grammar model. It was assumed that complex
grammatical transformations required transformational time to
reduce them to their kernel sentences. The greater intra-
and inter-sentence intervals in the slow one-trial condition

allowed.§§.the necessary transformation time.

111

ACKNOWLEDGEMENTS

The author wishes to express his gratitude to his
advisor and thesis director, Doctor CleSSen J. Martin, for
his scholarly contributions and constant encouragement in
the preparation of this dissertation. Special thanks are
also given to Drs. Lee S. Shulman, Robert L. Green, and
Robert H. Davis, for serving on the committee.

The author also wishes to thank his Sponsors, the Hazen
Foundation, for their support during the entire doctoral
program. Appreciation is also expressed to the Rockefeller
Foundation for the grant which made the completion of this

dissertation possible.

iv

Dedicated to the late James Aggrey of Africa

TABLE OF CONTENTS

CHAPTER I

Introduction

Statement of the problem

The Significance of the study
Theoretical Framework
Statement of Hypotheses

CHAPTER II

Review of Related Studies
Number of Presentations
Grammatical Transformations
Compressed Speech

Summary

CHAPTER III

Method
Stimulus Materials
Modes of Presentation
Massed Condition
Distributed Condition
Slow One-trial Condition
Subjects
Procedure
Individual Difference Measures

CHAPTER IV

Results
Pretest
Passage 1
A. Recall
B. Comprehension
Passage 2:
A. Rceall
B. Comprehension
Analyses of Correlations
Analysis of Individual Differences

CHAPTER'V

Discussion and Conclusions

vi

38
38

39

47
#9
52

58

Page

CHAPTER IV

Summary 66
APPENDICES 71
BIBLIOGRAPHY ' 118

vii

Table

10

11

12

13

14

LIST OF TABLES

Number of words, sentences and average
sentence length in the Low Difficulty
Versions and in the correSponding
transformed High Difficulty Versions

Amount of time taken to record each
version

The rate at which the passages and their
transformed high difficulty versions were
read on the tape-recorder

Number of subjects whose I.Q. and verbal '

reasoning ability scores were available

Means and standard deviations on the
pretest

Analysis of variance summary table of
performance of the six groups on the
pretest

Mean number of correct recall items and
standard deviations for each group on
Passage 1 ‘

Analysis of variance summary table of
recall on Passage 1

Differences between some specific pairs
of means of recall in Passage 1

Mean number of correct comprehension
items and standard deviation for each
group on Passage 1

Analysis of variance summary table of
comprehension

Differences between some specific pairs
of means of comprehension of Passage 1

Mean number of correct recall items and
standard deviations for each group on
Passage 2

Analysis of variance summary table of
recall

viii

Page

31

33

35

37

38

39

H0

40

#2

an

an

“5

u?

47

List of Tables (Continued)

Table
15

l6

17

18

19
20

21
22

23

Differences between some Specific pairs of
means of recall of Passage 2

Mean number of correct comprehension items
and standard deviation for each group on
Passage 2

Analysis of variance summary table of
comprehension

Differences between some specific pairs of
means of comprehension of Passage 2

Correlations between recall and comprehension

Correlations between I.Q. and recall of
Passages l and 2

Correlations between I.Q. and comprehension

Correlations between verbal reasoning ability
and recall

Correlations between verbal reasoning ability
and comprehension

ix

Page

48

50

50

51
53

5h
55

56

56

Figure

LIST OF FIGURES

Interactions between grammatical complexity
and modes of presentation in recall of
Passage 1

Interactions between grammatical complexity
and modes of presentation in comprehension
of Passage 1

Interactions between grammatical complexity
and mode of presentation in recall of

‘Passage 2

Interactions between grammatical complexity
and modes of presentation in comprehension
of Passage 2

Page

43

#6

“9

52

LIST OF APPENDICES

APPENDIX A

Pretest Passage

Passage 1: Low Difficulty Version

Passage 1: High Difficulty Version
Passage 2: Low Difficulty Version

Passage 2: High Difficulty Version

APPENDIX B

Questions for Pretest
Response Sheet for Pretest
Questions for Passage 1
Response Sheet for Passage 1
Questions for Passage 2
ReSponse Sheet for Passage 2

APPENDIX C

Raw Data

APPENDIX D

Individual Difference Measures

xi

Page

71
72

76
79

82
8Q
86
92
97
102

106

112

CHAPTER I
INTRODUCTION

One of the crucial issues in the psychology of learning
today is the extent to which number of trials or repetitions
actually influence learning. There are some learning theor-
ists who maintain that learning is a gradual process, such
that with increased number of trials or presentations there
should be obtained a negatively accelerated learning curve.
Learning theorists who hold this point of view are known as
continuity theorists.

Opposed to the continuity theory are those learning
theorists who adhere to the notion that learning is an all-or-
none affair. These learning theorists are referred to as non-
continuity theorists. They argue that in a learning situation,
a1§ makes certain hypotheses such that the hypothesis that
leads to the correct solution of a given problem is not
necessarily dependent on prior irrelevant hypotheses. Thus,
according to this view, number of trials or presentations is
not a significant factor in learning.

Although there exist these diametrically opposed views
on the role of repetition in learning, most learning psycho-
logists and progressive educators inherently accept the

notion that learning is a gradual process and that with more

repetitions or presentations of a given stimulus material.

a negatively accelerated learning curve should be obtained.
However, most of the studies which show a negatively
accelerated learning curve to be a function of number of trials
or_presentations are based on relatively simple learning tasks
(e.g.. nonsense syllable learning. serial learning. and paired-
associate learning). Consequently, the extent to which the
findings of these studies can be generalized to other kinds

of learning such as prose learning requires examdnation.

In prose learning, unlike the learning of simple tasks,
there are complex variables known as invisible strings (Flesch.
l9n6). According to Flesch, these invisible strings influence
how well a prose passage is understood. He points out that
in a simple short sentence pattern, there are invisible
strings; such that if_a sentence goes beyond the subject-
predicate-object type, it is more than likely to have a net
of criss-crossing strings that must be untangled before the
message it conveys is understood. For example. in a simple
sentence like "Robinson likes Florence" there is one string
between."Robinson" and "likes" and another between "likes"
and "Florence." But in a sentence that goes beyond the
subject-predicate-object there are many more such criss-
crossing strings. An example of such a sentence is given
below:

Here is Edmund Burke. the eminent British Liberal,

than whom.no European statesman was more horrified
by the outrages of the French Revolution (Flesch.

191,69 P0 32) e

According to Flesch. the above sentence is a tangle
because the clause is tied to the main sentence by "whom"
from which an invisible string leads to "Burke." But to reach
“whom" the word "than." which is tied to "more horrified."
must be jumped over. Flesch. therefore, recommends that the
sentence should be revised to read as follows:

No other European statesman was more horrified

by the outrages of the French Revolution than

Edmund Burke, the eminent British Liberal (Flesch,

1946. p. 32).

While Flesch's theoretical position has received support
from.some empirical studies (Coleman, 1962). it should be
pointed out that other studies examining the effects of
grammatical transformations clearly indicate that there are
more refined techniques for manipulating grammatical complexity.
For example, there is some empirical evidence (e.g.. Coleman,
1965) that active sentences are learned better than their
transformed passive versions. that non-embedded sentences
(e.g.. The cat killed the rat that ate the malt) are learned
better than embedded sentences (e.g.. The rat that the cat'
killed ate the malt), and that nominalized sentences (e.g..
The policeman's investigation of the incident involved hours
of tracing down suspects) are harder to learn than their de-
transformsd active versions (e.g.. When the policeman investi-
gated the incident. it involved hours of tracing down suspects).

Although these grammatical transformations indicate the
importance of certain transformational variables. the full

implications of this have not been thoroughly examined.

Moreover. in some studies, recall and comprehension are con-
founded (e.g.. Coleman, 1962). From a conceptual point of
view, recall and comprehension are not necessarily the same.
Recall of verbal materials may involve identification of
certain.previously stated facts or statements without necess-
arily understanding what these facts or statements mean or
imply. Thus, recall of factual information may be regarded
as verbatim learning (Cofer. 1941). Comprehension. on the
other hand, involves more complicated cognitive processes,
such as the derivation of some meanings that are not explicitly
stated, the drawing of some implications from the given
information, and also applying these derived meanings and
implications to other similar situations. Thus. comprehension
may be regarded as logical learning (Cofer, l9hl).

The above distinction between recall and comprehension
implies that recall of factual information is a necessary
condition for comprehension to occur, but the availability
of this factual information does not necessarily imply the
presence of comprehension. Thus, with exception of a few
cases (Wiener and Cromer, 1967, p. 622) there cannot be
comprehension without recall. although there may be recall
without comprehension.

In conclusion. perhaps, it would be appropriate to state
that recall of factual information is to the statement "we
learn by doing" what comprehension is to the statement "we
learn by thinking about what we are doing."

Although recall is related to comprehension, the extent

of the relationship is not known. In addition. little is
known about the factors which influence each or both of these
processes. More specifically. little research is available
which has examined the effects of mode of presentation and
grammatical complexity upon the recall and comprehension of

prose material.

Statement of the Problem:

The present problem.is, therefore, concerned with the
kinds of manipulations which possibly influence the recall
and comprehension of a prose passage. This study will examine
how number of presentations and rate of presentation influence
recall and comprehension of low difficulty and high difficulty

tape-recorded prose messages.

The Biggificance of the Study:
The significance of this study lies in the fact that the

results may provide some additional data on the factors
involved in the recall and comprehension of connected dis-
course of varying grammatical complexity. .A knowledge of

the role of repetitions and transformational time on recall
and comprehension of prose materials may be of significance
for prose learning theory development. and may also contribute
to the facilitation of school learning.

Theoretical Framework:
Some theoretical positions have been advanced to explain

memorization of sentences of varying grammatical transformations.

For example, one position.(Savin and Perchonock. 1965)
assumes that sentences with several grammatical transformations
occupy more space in immediate memory than do others (the ones
with fewer grammatical transformations). .Since each of these
grammatical transformations is encoded separately and also
occupies a characteristic amount of space in immediate memory.
it should be easier to encode sentences with fewer grammatical
transformations than sentences with more grammatical transforma-
tions. This being the case. it should be expected that
sentences with fewer grammatical transformations will be easier
to remember than sentences with more grammatical transformations.
Another theoretical position (Miller. 1962) assumes that
when.§§ are asked to learn sentences in an experimental
situation. they usually remember complex sentences in their
kernel forms. However. when the S5 are asked to recite the
original sentence exactly. they supplement their memory of
the kernel with a footnote about the syntactic structure.
Although these two theories would explain why recall of
low difficulty prose material should be easier than recall
of high difficulty prose material. it should be made quite
clear that the dependent variables measured in sentence learn- ‘
ing and in prose learning, though similar. are usually not
the same. In sentence learning the §,is usually required to
memorize the sentences and recite them exactly. but in prose
learning the §_is usually required to remember specific facts.
However, granted that the transformational processes postulated

by Miller occur in prose learning. it should be expected that

.§§ given prose material with more grammatical transformations
would have considerable difficulty in recall of certain factual
information. This is simply because thed§§' involvement in
the transformation of complex sentences into their kernel forms
while at the same time encoding them may interfere with the

§§f immediate recall of factual information. This line of
reasoning is based on the results of a study by Miller (1962).
The basic assumption in this particular study was that §§
would require a longer time to learn a more complicated
grammatical transformation than they would require to learn

a less complicated grammatical transformation. The test
consisted of eighteen kernel sentences, plus corresponding

sets of eighteen sentences which were produced from these ker-
nels by negative, passive, and passive-negative transforma-
tions. The fig were required to pair two of these four sets

of eighteen sentences (e.g., matching passive sentences with
their corresponding passive-negative forms).

The obtained results indicated that the passive trans-
formation or its inverse, the negative, and the passive
transformations took more time than the negative transforma-
tion or its inverse.

Since recall and comprehension are not necessarily
dichotomous, it would be necessary to examine some two similar
theoretical positions dealing with sentence comprehension.

One of these theoretical positions explains why nominalized
sentences are harder to comprehend than their active-verb

transformations. It postulates that nominalized sentences do

not have many specific references that exist in their active-
verb transformations although these references are implied
within the nominalized sentences. On the other hand, the
active-verb transformations express these Specific references
explicitly. Thus, since knowing these Specific references
is important in understanding the sentence. and since it is
difficult to deduce these Specific references from the given
nominalized sentence, it should be expected that nominalized
sentences will be more difficult to comprehend than their
active-verb transformations (Jesperson. l92h: Coleman, l96h).
The other theoretical position of sentence comprehension
states that in a simple short sentence pattern there are in-
visible strings, such that if a sentence goes beyond the
subject-predicate-object type. it is likely to have a net of
crisscrossing strings that must be untangled before the
sentence is understood. Consequently, for effective
communication. short sentences should be used. This emphasis
on the use of Short sentences is based on the assumption that
the use of short sentences gives the reader or the listener
ample opportunities "for breathing Spells and does not get
caught in invisible strings between words" (Flesch. 1946).
Since comprehension of a complicated sentence involves, first
of all. the untangling of these invisible strings, it
logically follows that the reader or listener who is not
caught in invisible strings between words should comprehend
the given information better than the reader or the listener

who is caught in these invisible strings.

A careful examination of the two theories of sentence
comprehension.clearly shows that the theory about nominalized
sentences, though quite relevant. would be inadequate in
explaining comprehension of prose material of various
grammatical transformations, because it is practically im-
possible to find a prose passage that consists of nominalized
sentences only. However, the theory would make a good
supplementary theory of prose learning only and only if the
prose material in question consists of some nominalized
sentences.

Flesch's theory too is quite relevant, but it is confined
to sentence length. and therefore would not explain comprehen-
sion of prose material that consists of other grammatical
transformations (e.g.. passives, etc.). Perhaps. a grammatical
transformation theoretical model that integrates Miller's
transformational model. nominalized sentence learning theor-
etical model. and.Flesch'S theoretical model would constitute
a good theoretical model of prose comprehension. Consequently.
such an integrated grammatical transformational theoretical
model of prose comprehension will be assumed here.

Since the mode of presenting low difficulty and high
difficulty messages is more than likely to have some
considerable effect on recall and comprehension of these
messages. it should be appropriate to examine how the suggested
three modes of presentation (massed, distributed. and slow one-
trial) are likely to influence recall and comprehension of low

difficulty and high difficulty tape-recorded messages. Before

10

doing so, however. it should be pointed out that there is

some empirical evidence showing that prose materials can be
presented at fairly rapid rates without loss of comprehension.
For example. in one study by Foulke, a passage was read at
three different rates (149. 16h.6, and 195.? wpm). This
passage was compressed to 275 and presented to one of the three
comparable groups of listeners. The results showed that the
groups did not differ Significantly on a comprehension test
(Foulke, 1966. p. 6).

An.examination of the suggested three modes of presenta-
tion Shows that only massed and distributed conditions have
been.intensively employed in simple learning studies. These
studies show that learning is better under distributed
practice condition than.under massed practice condition (Large.
1930: Mientzle, l9fl6: Duncan. 1951: Deese, 1958, pp. 187-19“;
Koppel. 196“: Hall. 1966, pp. 377-385: Underwood. 1966. pp.
569-570).

Two theoretical positions have been advanced to account
for this finding (Deese, 1958. pp. l9#-197). One of these
theoretical positions assumes that in.a distributed learning
condition. the §§ rehearse the stimulus material during the
rest periods. while in.a massed learning condition. the SS
do not rehearse. Consequently; the superior performance
obtained under distributed practice condition is attributed
to rehearsal during the rest periods. The other theoretical
position assumes that in a learning task, reactive inhibition
builds up, such that the organism or the §_learns not to

11

respond. This reactive inhibition is assumed to dissipate
during rest periods. and it is this dissipation of inhibition
that accounts for the fact that performance is better under
distributed practice condition than under massed practice
condition.

However. neither of these two theoretical positions are
readily applicable to recall and comprehension of prose
material. under massed and distributed conditions. There are
some obvious weaknesses in both theoretical positions. In
the first place; rehearsal has been controlled in some more
basic studies in.verbal learning. and learning is still found
to be better under distributed condition than under massed
condition.(Deese. 1958). Moreover. this phenomenon has been
shown to occur in animal learning where chances for rehearsal
are very slender (Bunch and Magdick. 1933: Holland. 1953;
Deese. 1958). The second theoretical position is also weak
on the ground that the assumption concerning conditioned
inhibition and dissipation of inhibition has not been
empirically substantiated with complex stimulus materials.

Perhaps, the best theoretical model that could explain
the better performance under distributed condition is the
grammatical transformational model. The application of this
theoretical model would necessarily assume that distributed
condition offers ample opportunities or time during the rest
periods (inter-sentence interval) for transforming complex
sentences into their kernel forms, whereas such opportunities

are lacking in a massed condition. In short, encoding of

12

high difficulty material should be better in a distributed
condition than in a massed condition. Since encoding of
high difficulty material is poorer in a massed condition.
it should be expected that this will lead to production
deficiencies. These production deficiencies should lead to
poorer recall and comprehension. This line of theorizing is
supported by the findings of the already cited Miller's
sentence matching study which showed thatI§§ required a
longer time to recognize a more complicated transformation
than they would require to recognize a less complicated
grammatical transformation (Miller, 1962).

Since no transformations are required in the low
difficulty material. it should be expected that recall of low
difficulty material should be better than recall of high
difficulty material. under a massed condition. High difficulty
material requires transformation time (inter-sentence interval),
but this transformation time is not available in the massed
condition. It should also be expected that recall of high
difficulty material under distributed condition. which allows
transformation time (inter-sentence interval), will be better
than recall of high difficulty material under a massed
condition. Since in the slow one-trial condition there exist
two types of transformational times (i.e.. inter-sentence
interval and intra-sentence interval) it Should be expected
that recall of high difficulty material will be better under
a slow one-trial condition. Lastly, recall of low difficulty
material should be better under a massed condition than

13

under a slow one-trial condition. This is because the low
difficulty material does not require transformation time;
hence hearing prose material twice Should lead to better re-
call of factual information in the low difficulty material.

With regard to comprehension, it should be expected that
comprehension of low difficulty material will be better under
a massed than under a slow one-trial condition. And
comprehension of low difficulty material should be better
than comprehension of high difficulty material. under a massed
condition. A Slow one-trial condition Should lead to better
comprehension of high difficulty material. Lastly.
comprehension of high difficulty material under distributed
condition should be expected to be better than comprehension
of high difficulty material under a massed condition.

Thus, operating within the transformational theoretical

framework, the following formal hypotheses are presented:

Statement of Hypotheses
A. £292.11:

Hypothesis 1. Modes of presentation and
grammatical complexity will
interact.

Hypothesis 2. Recall of low difficulty
material will be better than
recall of high difficulty mater-

ial, under a massed condition.

Hypothesis

B. Comprehension:

Hypothesis

Hypothesis

Hypothesis

Hypothesis

Hypothesis

3.

l.

3.

1%

Recall of low difficulty
material will be better under

a massed condition.

Modes of presentation and
grammatical complexity will
interact.

Comprehension of high difficulty
material will be better under a
slow one-trial condition than
under either a distributed or a
massed condition.

Comprehension of high difficulty
material will be better under a
distributed condition than under
a massed condition.
Comprehension of low difficulty
material will be better than
comprehension of high difficulty
material. under a massed condi-
tion.

Comprehension of low difficulty
material will be better under a

massed condition.

CHAPTER II
REVIEW OF RELATED STUDIES

The related research will be reviewed under the following
three headings: Number of Presentations, Grammatical Trans-

formations, and Compressed Speech.

1. Number of Presentations:

The major study examining the role of repetition in the
learning of prose material was conducted by Slamecka (1959).
The main objective of Slamecka's study was to determine
whether or not the recall of a passage of prose would increase
as the degree of its original learning was increased.
Secondly, the study also aimed at finding out whether or not
the curve of trial-by-trial acquisition would be negatively
accelerated, as is the usual case in Simpler learning tasks.

There were six groups of §§m Each group was given a
different number of presentations of the original passage
and three presentations of an interpolated passage. This
interpolated passage was used to control the experience of
all SS during the interval between original learning (0L) and
delayed recall (DR). Degree of original learning was varied

by the number of presentations of the passage (1, 3, 5, 7 and 9).

The passages were presented on a tape-recorder as follows:

original presentation, a single original recall, three

15

l6

interpolated presentations, a Single interpolated recall, and
a delayed recall of the original passage.

The S3 were told that they would be presented with
passages and that they would be required to memorize them.
There was a 5-Second interval between successive readings.

At the end of the listening period, the §§_were asked
to write down as much as they could recall of the first passage.

Results showed that the mean scores among the groups on
OL differed significantly. This showed that the number of
presentations influenced the level of acquisition. Plotting
of these acquisition means as a function of the number of
presentations generated a very smooth negatively accelerated
curve, consistent with the usual findings obtained with
unconnected material.

In another experiment, Slamecka (1959) tested the hypo-
thesis that recall of a prose passage would decrease as the
degree of learning of an interpolated passage increased.

The ngfor the experiment were divided into 5 groups,
every group being given a different number of presentations
of an interpolated passage, and three presentations of the
original passage.

The results showed that forgetting is directly related
to the degree of interpolated learning.

In.a subsequent study, Slamecka (1960) was interested
in investigating whether varying degrees of practice would
have the same effect on the recall of connected discourse as

on unconnected material. It was hypothesized that the rote

17

of retention of a prose passage is subject to retroactive
inhibition (RI) and that its recall is a direct function of
amount of practice in original learning (0L) and an inverse
function of amount of practice in interpolated learning (IL).
Subjects learned passages of connected discourse.
Three levels of OL practice (2, 4, and 8 trials) and three
levels of IL practice (0, h, and 8 trials) were given.
Results showed that OL and IL were a function of number
of practice trials. Recall of prose was subject to signi-
ficant RI and also varied directly with 0L practice and
inversely with IL practice.

2. Grammatical Transformations:

Recently, there has been an increasing number of studies
on grammatical transformations. However, only a few of
these studies are closely related to the present study.
Consequently, only these related studies will be reviewed.

To begin with, it would be necessary, perhaps, to point
out that although Flesch (l9u6, p. 33) emphasizes that for
effective communication we Should write short sentences, he
does not clearly Specify how much we improve comprehensibility
by shortening sentences. Consequently, this has been one of
the major problems to be investigated empirically. For
example, Coleman (1962) selected three difficult passages from
The Human Senses of Geldard and performed the following pro-
cedures: First, each passage was adjusted to contain 232 words,

#05 syllables, 32 prepositions, #5 "direct words," and lo

18

sentences; second, each passage was written into other versions
such that one version contained 6 sentences, and one contained
15 sentences, and the original one had 10 sentences. And
lastly. in each passage the first version had 15.4 words to

a sentence, and in the second they averaged 23.2, and in the
third they averaged 38.8.

Readability was measured by cloze tests given immediately
after the §.had finished reading each passage. Mean number
of words correctly inserted per subject were 22.4, 21.4, 20.9
for the words for passages of 15.h, 23.2 and 38.6 words to a
sentence reSpectively. These differences are statistically
Significant and a good indication that technical passages
divided into short sentences were significantly more
comprehensible than their long counterparts. The magnitude
of the improvement in.comprehensibility due to shortening
sentences was found to be about 6 per cent.

In a subsequent study, Coleman and Blumenfeld (1963)
investigated the relative comprehensibility of different
grammatical transformations of a single sentence. The trans-
formations compared in this particular study were nominalize-
tions derived from.verbs (e.g., John's description of it) and
their grammatical transformations using the active verb (John
described it).

Each sentence was written in two versions (nominalized
and active verb), and for eachԤ half of his sentences were
in the active-verb version and the other half were in the

nominalized version. The results showed that the performance

19

on the active verb version was better than the performance
on the nominalized version.

In another similar study, Coleman (196“) carried out four
experiments. In the first experiment three transformations
were applied simultaneously to difficult technical prose
(nominalizations, passives and adjectivalizations were trans-
formed to their active-verb counterparts, and the original
versions were compared to the transformed versions for
readability). The simplified versions were comprehended
better than the original versions.

In the second experiment, each subject read four passages
in.the original version and four in the simplified version.
He was then asked to write each of these passages exactly,
immediately after reading each of them. The passages were
simplified by applying the same three transformations used
in the first experiment. The results showed that the
simplified versions were recalled better than the originals.

In the third experiment, 20 nominalized sentences were
compared with their active-verb transformations. The measure
of learning was the same as in the second experiment. The
results again showed the active-verb transformations were
recalled better than the nominalized versions.

In the last experiment, 10 nominalized sentences were
compared with their active-verb transformations in terms of
the amount of time required to learn each version. Sentences
were presented tol§§ on a memory drum at the rate of one

second. The measure of learning was in terms of trials to

20

perfect recall. As it was expected, active-verb transforma-
tions took less time to be learned than the original
nominalized sentences.

Coleman (1965) further examined different categories of
grammatical transformations. Four experiments were carried
out. The first experiment contained the examination of
nominalizations, and the second compared actives to passives.
In the third experiment, adjectivalizations (e.g., the urgency
of immediate demands is allowed to usurp attention) were
compared to their detransformed adjective versions (e.g.,
Immediate demands are so urgent that they are allowed to usurp
attention). And lastly, a comparison of embedded sentences
(e.g., The rat that the cat killed ate the malt) with non-
embedded sentences (e.g., The cat killed the rat that ate the
malt) was made.

The results of the four experiments showed that the de-
transformed active-verb versions, measured by a cloze test,
were easier to learn, actives were better retained than
passives, and that performance on nonembedded.versions was
better than performance on embedded versions. There was no
indication that detransforming any and all adjectivalizations
will have significant effect in making prose easier to
comprehend.

Mehler (1963) also carried out a similar study to those
by Coleman. He presented a set of eight short sentences to
the S'who was instructed that he would be tested for recall.

The sentences differed in their grammatical forms (e.g., active

21

or passive, affirmative or negative, declarative or interrog-
ative).

The eight kernel sentences and their transformations gave
6% sentences which were divided into 8 groups of 8 sentences
in such a way that in each group no sentence and no trans-
formation was represented more than once. The results showed
that kernel sentences were learned with greater facility.

The relative effects of these grammatical transformations
on Speed of understanding has also been investigated in a
study by Gough (1965). In this particular study, 8 descriptive
kernel sentences, plus the negative, passive and.negative-
passive transformations were used. There were, thus, 32 sen-
tences: and each of these 32 sentences could be paired with
any of the 16 experimental events which confirmed or falsified
the sentences.

The obtained results indicated that active sentences were
verified faster than passive sentences, affirmative sentences
faster than.negative sentences. The results also Showed that
true sentences were verified faster than false, and truth
value interacted with the affirmative-negative variable.

Slobin (1966) in a similar study tried to examine the
degree to which children and adults were able to verify kernel.
passive, negative, and passive-negative sentences. The method
employed in this study was as follows: a sentence was read
and then a picture was exposed. The §.was required to say
whether the sentence was true or false in relation to the

exposed picture.

22

The results showed that female §§ of all ages tended to
respond more rapidly than.males. The main effect of grammar
was found to be significant, and consistent with the previous
findings, kernel sentences required less time to verify than
passive sentences, and passive sentences required less time
than.negative sentences, and negative sentences less time than
passive-negative sentences. This was true at all age levels
(kindergarten, second, fourth and Sixth grades, and adults).

The results also showed that true negatives (e.g., The
cat is not chasing the dog) were more difficult to verify
than false negatives (e.g., The dog is not chasing the cat).
There was also some good evidence to show that reversible
sentences (e.g., The dog is chasing the cat) were harder to
verify than non-reversible sentences (e.g., The girl is water-
ing the flowers). This finding hold true at all age levels.

Age was also found to be a significant variable in that
both reaction time and errors in verification of the sentences
tended to diminish with age.

In a similar study, Smith (1965) attempted to examine
whether the time involved in making passive or negative trans-
formations (or their inverse) is independent of their effect
on the meaning of the transformed sentence.

Pour tasks, each requiring the conversion of a list of
16 sentences according to rules involving either a negative
or passive transformation (or their inverse) and resulting
in the reversal or nonreversal of sentence meaning were

employed in the study. The task rules also involved the

23

changing of words within each sentence to opposite or
synonymous form. A Single negative transformation or word
change to the opposite reversed the meaning of a sentence,
while a Single passive transformation or word change to
synonymous form left the meaning unchanged. Two meaning
reversals within the same sentence cancelled each other and
left the total meaning unchanged.

Group I §§ performed tasks involving the following trans-
formations: 1) one transformation and one word change, the
conjunction of negative transformation and one word change to
the opposite form: 2) the conjunction of negative transforma-
tion.and word change to opposite or of passive transformations
and word change to synonymous form.

Group II §§ performed similar tasks, each involving one
transformation and two word changes. Thus, for each group
of RE: the tasks resulted in a reversal of sentence meaning
and two did not.

The results Showed that change of sentence meaning
resulted in longer performance times in tasks involving passive
transformations, but not in tasks involving negative
transformations.

In.a somewhat related study, but performing different
kinds of transformations in a prose passage, Martin and Alonso
(1967) examined whether or not high informational, condensed
verbal materials can be comprehended as well as, if not better

than, full length text verbal materials.

24

The §§ for the study were blind children. One-third of
the SE read a traditional passage, one-third read a medium
length version, and the remaining one-third read a short
telegraphic version. All three versions contained the same
essential information.

The following three conditions were employed to obtain
information acquired under each of these three versions: 533
relations questions, which were designed to measure s3}
retention and distortion of the set relations: multiple-choice

uestions, which tested the retention of factual type
information: and reconstruction, in which.§§ were required
to reconstruct the story from.memory.

The obtained results showed that the SE who read the
medium and Short versions required significantly less time
than.§§,who read the traditional version.

Some studies in this general area have also attempted to
develop theoretical models to account for the learning of
sentences of various grammatical transformations. For example,
a study by Savin.and Perchonock (1965) attempted to examine
why some of these grammatical transformations are easier to
remember than others. More specifically, the purpose of the
study was to measure the amount of immediate memory that is
occupied by several grammatical features. The grammatical
features of interest were the negative, passive, question,
negative-question, emphatic, and wh-transformations.

The investigators presumed that each of these features

is encoded separately and therefore each occupies a

25

characteristic amount of space in immediate memory. It was
postulated that the sentences with several transformations
occupy more Space in immediate memory than others (the ones
with fewer transformations). The problem was, therefore, to
measure the amount of Space remaining in memory after a
sentence had already been encoded. Consequently, this
additional Space was measured by seeing how much additional
material could be remembered along with the sentence. The
rationale here was that if a sentence filled most of the
memory. little or nothing extra can be remembered along with
it. On the other hand, if another sentence was encoded
compactly. then there would be Space for several other items
in addition to it.

The S was therefore given a sentence followed by a string
of words and was required to recall as many as possible of
the extra words. The number of extra words recalled was taken
as a measure of the memory capacity remaining after the sen-
tence had been encoded.

The results supported the assumption that sentences with
more transformations occupy more space in immediate memory
than do sentences with fewer transformations. However, it is
also possible that during the recall the Slgas still altering
the syntactic form of the sentences with more transformations.
This process of alteration should interfere with the number
of extra words that can be recalled. This line of argument
is based on.Miller's (1962) theoretical position that what

s3 remember is the kernel sentence. but when asked to recite

26

the original sentence exactly, they supplement their memory
of the kernel sentence with a footnote about the syntactic
structure.

Although most of the experimental and theoretical studies
on grammatical transformations have primarily been by psycho-
logists, the work of the linguist Chomsky (1968) has also
contributed to our theoretical understanding in this area.
According to Chomsky, people are born with prediSpositions
to use a language in a certain way. such that if a sentence
whose surface structure does not correspond to its deep
structure (the way people are predisposed to use a sentence),
certain grammatical transformations must be made for the
sentence to be understood. Chomsky's theorizing has influenced
the work of psychologists such as Miller (1962) and Savin and
Perchonock (1965).

3. QQEBressed Speech:

There is an increasing amount of research being conducted
on time compressed speech. Time compressed speech essentially
means that if a Speech has been delivered at a normal word
rate (175 wpm) the same speech can be accelerated by increas-
ing word rate (Foulke, 1966).

The dependent variables studied in time compressed
speech are usually intelligibility and comprehension. In one
study (Garvey. 1953) 95 per cent or greater intelligibility
was obtained when the original Speed had been accelerated 1.5.

1.75. and 2.0 times. Accelerating original speed 1.67 and

27

2.5 times gave mean intelligibility scores above 90 per cent.

Accelerating the original speed 3.0 times produced a mean
intelligibility score of 78 per cent. When the acceleration
of the original speed reached 3.5 times, intelligibility
dropped to 50 per cent. and the mean intelligibility score’
did not drOp below the 50 per cent level until the original
speed had been accelerated u.0 times.

These findings were supported by the findings of a sub-
sequent study by Foulke and Sticht (Foulke. 1966, p. 26)
which showed that there was a 9 per cent loss in intelligibility
at #25 wpm. Compressing Speech to approximately twice the
normal rate (175 wpm) led to only six per cent loss in
intelligibility.

However. comprehension was affected much more than
intelligibility. Comprehension begins to decline at 275 wpm.
This is well substantiated by the finding of a study (Foulke.
1966) in which a passage was read at three different rates
(1&9, 16%.6. and 195.7 wpm) and then compressed to 275 wpm.
The compressed passage was presented to one of the three
comparable groups of listeners. The results showed that the
groups did not differ significantly on a comprehension test.

In.a similar study (Orr. Friedman and Williams. 1965) SE
were given systematic practice in listening comprehension of
prose material tape-recorded at progressively increasing rates
of 325 wpm to #75 wpm. Again the results showed that increas-
ing Speed of presentation in words per minute up to double
the normal 175 wpm rate did not produce a significant loss in

28

comprehension.

The findings of the studies cited here are consistent
with the findings of other studies in this general area of
time compressed speech (Fairbanks. Guttman.and Myron. 1957:
Foulke, Amster. Nolan and Bixler. 1962) which indicated that
there is very little loss in comprehension until the rates

approach or exceed 300 wpm.

Summary:
Studies cited here clearly indicate that the role of

repetitions in prose learning has not been thoroughly examined.
Although Slamecka (1959 and 1960) has attempted to do this,

the relative effects of number of presentations on prose
learning when time is held constant are not known.

Also the effects of availability of transformational time
on recall and comprehension of prose material has not been
empirically examined.

Thus. this study is designed to examine the effects of
repetitions on recall and comprehension of prose materials of
varying grammatical complexity when time is held constant.
Secondly. it is necessary to examine the effects of providing
transformation time (inter-sentence interval and intra-sentence
interval) on recall and comprehension of low difficulty and
high difficulty prose passages.

Operating within the grammatical transformational theor-
etical framework discussed in Chapter I. it is predicted that
number of presentations will be a significant factor in the

29

recall and comprehension of the low difficulty material. while
transformational time (inter-sentence interval and intra-
sentence interval) will be a significant factor in the recall
and comprehension of high difficulty material.

Details concerning the development of the stimulus
material. modes of presentations and other related experimental

procedures are described in the next chapter.

CHAPTER III
METHOD

_timulus Materials:

A 35A-word standardized reading passage for grade nine
was selected from "Tests ofgcademicgrogress: Forml"

(p. 54). The average sentence length of this passage was
16.8 words.

A second standardized reading passage (655 words) was
selected from."Iowa Tests of Basic Skills: Form 3" (p. 22).
The average length of this passage was 13.9 words. The two
passages, in their standardized forms, constituted the L2!
Difficulty Versions.

The following procedures were employed in transforming
these Low Difficulty Versions (LDV) into High Difficulty
Versions (HDV): Wherever possible, sentences in the LDV were
changed into passives, noun clauses, embedded versions.
nominalized versions. and wherever possible. combining these
transformed versions into long complex sentences. This
procedure is justified on the ground that sentence learning
research literature reviewed in Chapter II clearly indicates
that these grammatical transformations are difficult to
recall or comprehend.

An example of LDV and its transformed HDV are given below:

30

31

Low Difficulty Version

The paintings of ancient
peoples tell us of their prob-
lems and their heroes. From
their art we can learn about
their government and their
religion. For example, we
know that the Egyptians had
powerful and glorified rulers
and their religion included
included a belief that there
was life after death.

Table 1.

High Difficulty Version

By the paintings and art
of ancient peoples we are
told of their problems and
heroes, their government and
religion. That the Egyptians
had powerful and glorified
rulers, for example, and
that included in their
religion was a belief that
there was life after death,
we know.

Number of words, sentences, and average sentence

length in the Low Difficulty Versions and in the Corresponding
transformed High Difficulty Versions.

 

 

 

 

 

Words Sentences Average
Words/Sentence

Passage 1.

LDV 354 21 16.8

HDV 376 13 ' 28.9
Passage 2.

LDV 655 47 13. 9

HDV 754 22 34.2

 

Ten recall and ten comprehension questions were obtained

for each of the two passages.

Recall questions were designed

to test the S's ability to recall factual information, while

comprehension questions were designed to test the S's ability

to organize and understand given or implied information.

32

Some examples of recall and comprehension questions are given

 

below.
Recall Comprehension
A complex set of written symp Which of the following would
bols developed by the Egypt- make the best title for this
ians were known as selection?
1) Braille l) Custer the Great
2) Microscopics 2) The Yellow Hair
3) Hieroglyphics 3) The military career of

George Armstrong Custer
4) Cuineform
4) The death of Custer
5) None of these

According to the selection.

What is the Little Bighorn? what is the principal advant-
age of water color over other
1) An Indian reservation art mediums?
2) One of the Black Hills 1) A wider variety of colors
is possible
3) A river

2) Cheaper materials are
4) A pass between two hills available_

3) Larger surfaces can be
covered more easily

4) All the above are advantages
5) The selection gives no clue
Modes of Presentation:
There were three modes of presentation: massed, distri-
buted, and slow one-trial.
Massed Conditign: The LD and HD passages were presented twice
at approximately 208 wpm. The second presentation occurred

immediately upon completion of the first presentation.

33

[gistributed Condition: AS in the massed condition, both the

LD and HD passages were presented twice at approximately 222
wpm. However. in this condition there was a 5 second pause

at the completion of each sentence.

§low One-trial Condition: The LD and RD passages were presented
only once at approximately 117 wpm.

Thus, there were six experimental conditions: LD-Massed,
LD-Distributed. LD-Slow One-trial. HID-Massed, HD-Distributed,
and HD-Slow One-trial.

Table 2 Shows the amount of time it took to tape-record

each of these conditions.

Table 2. Amount of time taken to tape-record each version.

 

 

 

 

Number of words minutes seconds

Passage 1:

LD-Massed 354 3 36

LD-Distributed 354 6 3O

LD-Slow One-trial 354 4 45

HD-Massed 376 3 “0

HD-Distributed 376 5 30

HD-Slow One-trial 376 4 20
Passage 2:

LD-Massed 655 6 4O

LD-Distributed 655 13 40

LD-Slow One-trial 655 9 30

34

Table 2: (Continued)

 

 

Number of words minutes seconds
PassageZ:
HD-Massed 754 ' 6 30
HD-Distributed 754 10 00
HD-Slow One-trial 754 7 50

 

 

The rate at which each of these conditions was tape-
recorded was calculated as follows: the total time taken to
record massed condition was halved and the total number of
words in that condition was then divided by this quotient.

In the distributed condition. the number of pauses in each
version was multiplied by five (i.e.. the 5-Second pause at
the end of each sentence). The resulting figure was sub-
tracted from.the total amount of time in that condition. The
remaining amount of time was halved. and the total number of
words in that condition was divided by this quotient. Since
in the slow one-trial condition. five seconds elapsed before the
next sentence could be read. this had to be accounted for when
calculating the reading speed. Consequently. the number of
pauses in each version was multiplied by five. The obtained
product was subtracted from the total amount of time for this
condition. and then the total number of words in this condi-
tion was divided by the remaining amount of time. Table 3
shows the rate at which the passages and their transformed

high difficulty versions were read on the tape-recorder:

35

Table 3. The rate at which the passages and their transformed
high difficulty versions were read on the tape-recorder.

 

 

 

 

 

 

Pasaage 1: Version Words per Minute
LD-Massed - 196.7
LD-Distributed 223.5
LD-Slow One-trial 118
HD-Massed 20509
HD-Distributed 214.9
HD-Slow One-trial 112.8

Passage 2:

LD-Massed 196.5

LD-Distributed 218.3

LD-Slow'One-trial 115

HD-Massed 232

HD-Distributed 232

HD-Slow One-trial izu
Subjects:

One hundred and twenty S5 were obtained from six classes
of grade nine students at Okemos High School. It was not
possible to randomly assign these §§ into six experimental
groups because the authorities felt that the students had
been initially randomly assigned to these classes. Secondly.
it was felt that random assignment of these gs into six
experimental groups would have seriously interfered with

class schedules.

36

Procedure:

Since it was not possible to randomly assign the SE into
six experimental groups, it was necessary to administer a
pretest to determine whether or not the six groups were similar
in prose learning ability. Consequently. a short passage was
selected from "Tests of Academic_Progress: Form 1" (p. 47)
for this purpose. This passage was read on a tape-recorder
at the normal speed. The §§.listened to the tape and then
took a Short quiz.

Each of the six groups was assigned to one of the six
experimental treatments (LB-Massed. LD-Distributed. LD-Slow
One-trial, HD-Massed. HD-Distributed. HD-Slow one-trial).
after the pretest.

At the beginning of each experiment. the s3 were told.
on the tape, that some passages were going to be read to them
on the tape, and that at the end of each passage. they would
be given a quiz.

After the §§_had finished taking the quiz on the first
passage. they were told to take a short break (5 minutes) and
then be prepared for another passage.

Performance on recall and comprehension was scored in

terms of items answered correctly.

Individual Difference Measures:
I.Q. (Lorge Thorndike) scores and Verbal Reasoning
Ability (Differential Aptitude Test) scores were available

for some of the S5. It was necessary to obtain this

37

incidental information to facilitate correlational analyses
between individual difference measures and the dependent
variables (recall and comprehension). Table 4 shows the
number of §§. in each of the six groups. whose individual
difference measures were available.

Table 4. Number of subjects whose I.Q. and Verbal Reasoning
Ability scores were available.

 

 

 

Group Number
LD-Massed 10
LD-Distributed 12
LD-Slow One-trial l7
HD-Massed l9
HD-Distributed l8

HD-Slow One-trial l6

 

 

CHAPTER IV
RESULTS
Pretest

The rationale for the pretest was to establish the
comparability of the groups to recall and comprehend prose
material. The number of correct items on the pretest was
recorded for each §. Table 5 presents means and standard

deviations on the pretest.

Table 5. Means and standard deviations on the pretest.

 

 

 

Group Mean SD

HD-Massed 2.85 1.49
HD-Distributed 3.15 p 1.49
HD-Slow One-trial 3.15 1.15
LD-Massed 3.10 1.84
LD-Distributed 3.15 1.88
LD-Slow One-trial 2.90 1.22

 

 

Table 6 presents the summary of a simple 1 x K

analysis of variance which showed that the six groups were

similar.

39

Table 6. Analysis of variance summary table of performance
of the six groups on the pretest.

 

 

 

Source SS df MS F
Between 1.90 5 A .38 .15
Within 283.80 114 2.48

Total 285.70 119

 

 

Thus, it could be assumed that the groups were compar-
able in their ability to learn from prose material. 0n the
basis of these results it can, thus, be assumed that
differences in the performance of these six groups in the
six different experimental conditions were due to the
experimental treatments rather than initial differences among
the groups.

Statistical analysis of performance on recall and

comprehension were performed separately for each passage.

Passage 1:

A. Recall:-
The total number of correct recall items was recorded
for each S. Table 7 presents the means and standard devia-

tions for the six groups.

40

Table 7. Mean number of correct recall items and standard
deviations for each group on Passage 1

 

 

 

Group Mean. SD
HD-Massed 3.45 F 1.36
HD-Distributed 4.80 1.72
HD-SIOW One-trial 6.20 2.40
LD-Massed 6.30 2.05
LD-Distributed 5.70 1.65
LD-Slow One-trial 6.15 1.46

 

 

Table 8 presents an analysis of variance which shows
that both level of difficulty and mode of presentation had
significant effects on recall. A test for homogeneity of

variances showed that the variances were similar (F-max =

3.11. p 7.05) 0

Table 8. Analysis of variance summary table of recall on
Passage 1.

 

 

 

Source SS df MS F P
Difficulty 45.63 1 45.63 13.26 .0005
Mode 35.82 2 17.91 5.20 .007
Interaction 43.72 2 21.86 6.35 .002
Error 392.30 114 3.44

Total 517.47 119

 

 

41

Since there was a significant mode x difficulty inter-
action. a test for simple main effects (Bruning and Kintz,
1968, pp. 117-123) of mode within each difficulty level was
performed. The null hypothesis of mode main effects within
the low difficulty material was not rejected. This indicated
that mode of presentation is not an important variable when
difficulty is low. The null hypothesis of mode main effects
within high difficulty level was rejected. Thus, mode of
presentation contributes to variability in recall of prose
material under high difficulty conditions. The locus of
the Significant main effect showed that recall of high
difficulty material under the distributed condition was
significantly better than recall of high difficulty material
under the massed condition: recall of high difficulty
material under the slow one-trial condition was significantly
better than recall of the high difficulty material under either
the distributed condition or the massed condition. Also,
recall of high difficulty material was significantly poorer
than recall of low difficulty material under the madsed
condition. Table 9 presents the results of the individual

comparisons.

Table 9.
of recall in Passage 1.

42

Differences between some specific pairs of means

 

 

 

 

Egggitlons HD-M HD-D LD-D LD-S HD-S LD-M
Means 3.45 4.80 5.70. 6.15 6.20 6.30

HD-M 3.45 --- 1.35** 2.75** 2.85**

HD-D 4.80 1.35** ~--- .90 1.40**

LD-D 5.70 .90 ---- .45 .60

LD-S 6.15 .45 ---- .05 .15

HD-S 6.20 ----

LD-M 6.30 ----

 

**Significant differences beyond .05 level

Key to abbreviations in Table 9:

HD-M = High Difficulty Massed

HD-D a High Difficulty Distributed
LD-D = Low Difficulty Distributed
LD-S aLow'Difficulty Slow One-trial
HD-S = High Difficulty Slow One-trial
LD-M = Low Difficulty Massed

All these hypotheses, except one, were confirmed by
these results. Only the hypothesis which had predicted re-
call of low difficulty material to be better than recall
of high difficulty material. under the distributed condition,
was not supported. However. there was a trend in the
predicted direction.

Figure 1 presents the interaction between levels of

difficulty and modes of presentation.

43

‘5 s 1004 Ohc—Tw'q‘

6‘

Mean number of correct recall items

 

 

5
‘ bl 5+“. Lufgd
4.
'4¢~95124
3 _
Low Type of Material High

Fig. 1. Interactions between grammatical complexity and
modes of presentation in recall of Passage 1.

B. Comprehension:-
The total number of correct comprehension items was

recorded for each S. Table 10 presents the means and standard

deviations for each of the six groups on Passage 1.

Table 10. Mean number of correct comprehension items and
standard deviation for each group on Passage 1.

44

 

 

 

Group Mean SD
HD-Massed 3.20 1.69
HD-Distributed 4.00 1.70
HD-Slow One-trial 5.10 1.37
LD-Massed 5.45 2.06
LDéDistributed 5.40 1.83
LD-Slow One-trial 5.55 2.36

 

 

Table 11 presents an analysis of variance which showed

a significant main effect due to difficulty. Mode of presenta-
tion was not significant, but again there was a trend in the
predicted direction. A test for homogeneity of variances

showed that variances were similar (F-max :- 2.96.. p.>.05).

Table 11. Analysis of variance summary table of comprehension.

 

 

 

Source SS df MS F P
Difficulty 56.03 1 56.03 15.37 .0005
Mode 20.42 2 10.21 2.80 .065
Interaction 16.22 2 8.11 2.22 .113
Error 415.70 114 3.65

Total 508.37 119

 

 

45

Individual means were compared to determine which ones
differed significantly from.each other. Tukey's significant
gap of 1.19 Showed that comprehension of low difficulty mate-
rial was significantly better than comprehension of high
difficulty material. under massed condition and distributed
condition: and comprehension of high difficulty material
under slow one-trial condition was significantly better than
comprehension of high difficulty material under massed condi-
tion (Table 12).

Table 12. Differences between some Specific pairs of means
of comprehension of Passage 1.

 

 

Conditions HD-M HD-D HD-S LD-D LD§M LD4S
Means 3.20 4.00 5.10 5.40 5.45 5.55

 

HD-M 3.20 ---- 1.9o** 2.25**
HDbD 4.00 .80 ---- 1.10 1.uo**

HD-S 5.10 ----

LD-D 5.40 ----

LD-M 5.45 ----

LD-S 5-55 --_-

 

**Significant differences beyond .05 level.

Thus. the results confirmed three hypotheses and not two.

The two hypotheses that were not supported had predicted

comprehension of high difficulty material under distributed

condition to be better than comprehension of high difficulty

as

material under massed condition, and an interaction between
levels of difficulty and modes of presentation.
Figure 2 shows these interactions between levels of

difficulty and modes of presentation.

5‘0“, One-‘lYtul

A!“ 1}.“

 

bls‘lviLu‘lgo‘

.4:

l
1 Massed

I»

 

Mean number correct comprehension items

7.:

 

Low High
Type of Material

Fig. 2. Interactions between grammatical complexity and
modes of presentation in comprehension of Passage 1.

4?
Passage 2

A. Recall:

The total number of correct recall items was recorded
for each.§. Table 13 presents the means and standard
deviations for the six groups.

Table 13. Mean number of correct recall items and standard
deviation for each group on Passage 2.

 

 

 

Group Mean SD

HD-Massed 4.55 1.69
HD-Distributed 5.25 1.6?
HD-Slow'One-trial 5.80 1.50
LD-Massed 6.40 1.83
LD-Distributed 6.40 1.69
LDmSlow'One-trial 6.00 1.14

 

 

Table 14 presents an analysis of variance which showed
that levels of difficulty had a significant effect on recall.

Table 14. Analysis of variance summary table of recall.

 

 

 

Source SS df MS F P
Difficulty 34.13 1 34.13 12.65 .001
Mods 4.12 2 2.06 0.76 .469
Interaction 13.72 2 6.86 2.54 .083
Error 307.50 114 2.70

Total 3590“? 119

 

 

48

Tukey's significant gap of 1.02 indicated that recall
of low difficulty material was significantly better than
recall of high difficulty material. under the massed condi-
tion, and the distributed condition. Recall of high
difficulty material under the slow one-trial condition was
significantly better than recall of the high difficulty
material under the massed condition. Table 15 presents the
differences between Specific pairs of means of recall of
Passage 2. The predicted interaction between levels of
difficulty and mode of presentation was not significant.
However. there was a trend in the predicted direction.
Variances were found to be similar (F-max I 2.58, p)v.05).

Table 15. Differences between some specific pairs of means
of recall of Passage 2.

 

 

Conditions HD-M HD-D HD-S LD-S LD-M LD-D

 

Means 4.55 5.25 5.80 6.00. 6.40 6.40

 

HD-M 4.55 ---- 1.25** 1.85**

HD-D 5.25 .70 ---- .55 1.15**
HD-S 5.80 ----

LD-S 6.00 ----

LD-M 6.40 ----

LD-D 6.40 ----

 

 

**Significant differences beyond .05 level

Figure 3 Shows these interactions between levels of difficulty

and modes of presentation.

Mean number of correct recall items

49

/

 

*3 Slave 0 ng—ivic‘al

bl 3+V; Lu*¢d

 

 

5.
Massed
4
15L
12 ._4
Low High

Type of Material

Fig. 3. Interactions between grammatical complexity and
mode of presentation in recall of Passage 2.

B. Comprehension

The total number of correct comprehension items was
recorded for each.§. Table 16 presents the means and

standard deviations for the six groups.

Table 16. Mean number of correct comprehension items and

standard deviation for each group on Passage 2.

 

 

 

Group Mean SD
HD-Massed 4.45 1.88
HD-Distributed 4.70 1.68
HD-Slow One-trial 5.70 1.23
LD-Massed 5.55 1-75
LD-Distributed 5.75 1,6h
LD-Slow One-trial 6.10 1.61

 

 

Table 17 presents an analysis of variance which Showed
that both levels of difficulty and modes of presentation had
significant effects on comprehension. A test for homogeneity
of variances showed that the variances were homogeneous

(F-max = 2.34, p>.05).

Table 17. Analysis of variance summary table of comprehension.

 

 

 

Source SS df MS F P
Difficulty 21.67 1 21.67 7.63 .007
Modes 17.55 2 8.77 3.09 .049
Interaction 3.05 2 1.52 0.54 .586
Error 323.85 114 2.84

Total 366.12

 

 

Tukey's Significant gap of 1.04 Showed that comprehension
of low difficulty material was significantly better than

51

comprehension of high difficulty material, under the massed
condition and the distributed condition: and comprehension
of high difficulty material under the Slow one-trial condi-
tion was Significantly better than comprehension of high
difficulty material under the massed condition. Table 18
presents the results of the Tukey test for individual
comparisons.

Table 18. Differences between some Specific pairs of means
of comprehension of Passage 2

 

 

Conditions HD-M HD-D LD-M HD-S LD-D LD-S

 

Means 4.45 4.70 5.55 5.70 5.75 6.10

HD-M 4.45 ---- 1.10** 1.25**

HD-D 4.70 .25 ---- 1.00 1.05**

LD-H 5-55 ----

HD-S 5.70 ----

LD-D 5.75 ----

LD-S 6.10 ----

 

 

**Significant beyond .05 level

Figure 4 shows the interactions between grammatical
complexity and modes of presentation in comprehension of

Passage 2.

Mean number of correct comprehension items

52

:3

 

~O$lew Oflfi-FV‘J

Uh

bl8+vlbu1~cal
Massed

4’

U”

 

Low High
Type of Material

Fig. 4. Interactions between grammatical complexity and
modes of presentation in comprehension of Passage 2.
Analyses of correlations between recall and comprehension.
under the six conditions

Correlational analyses between recall and comprehension
were done to find out to what extent recall and comprehension

are related, under each of the six conditions (Table 19).

53

Table 19. Correlations between recall and comprehension.

 

 

 

 

 

Group r P

Passage 1:

HD-Massed .09 I n.s.
HD-Distributed .07 n.s.
HD-Slow One-trial .21 n.s.
LD-Massed .67 .01
LD-Distributed .51 .02
LD-Slow One-trial .53 .02
Passage 2

HD-Massed .27 n.s.
HD-Distributed .49 .05
HD-Slow One-trial .21 n.s.
LD-Massed .39 n.s.
LD-Distributed .62 . .01
LD431ow'0ne-trial .54 .02

 

 

These correlations Show that recall and comprehension
of high difficulty material were not significantly correlated,
under the massed condition. distributed condition. and Slow
one-trial condition. On the other hand. recall and
comprehension of low difficulty material were significantly

correlated under these same conditions.

54

Anaiysis of Individual Differences:

Although the analysis of the individual difference
variables was not central to this study, such information
may have value in determining the extent to which these
variables were correlated in the various presentation condi-
tions. This information may be of value in Specifying the
extent to which individual difference variables predict recall
and comprehension of prose material under various modes of
presentation.

Table 20 presents the correlations between I.Q. and
recall of passage 1 and 2, under all six experimental conditions.

Table 20. Correlations between I.Q. and recall of passages
l and 2.

 

 

 

 

 

Passage 1
Group N r P

HD"Massed 19 01"]. 11.8.
HD-Distributed 18 .21 n.s.
HD-Slow One-trial 16 .75 .01
LD‘Massed. 10 .72 002
LD-Distributed 12 .56 .05
LD-Slow One-trial 17 .46 n.s.
Passage 2

HD-Massed 19 .28 n.s.
HD-Distributed 18 .38 n.s.
HD~Slow'0ne-trial 16 .53 .05
LD-Massed 10 .17 n.s.
LD-Distributed 12 .79 .01
LD-Slow'One-trial 17 .17 n.s.

 

 

These correlation coefficients indicate that I.Q. and

recall are significantly correlated under HD-Slow One-trial,

55

LD-Massed (except in Passage 2). and LD-Distributed conditions.
Table 21 presents the correlations between I.Q. and comprehen-
sion of passage 1 and passage 2. under all Six experimental

conditions.

Table 21. Correlations between I.Q. and Comprehension.

_:—

 

 

 

 

Group N r P

Passage 1

HD-Massed 19 .20 n.s.
HD-Distributed 18 .09 n.s.
HD-Slow One-trial 16 .10 n.s.
LD-Massed 10 .82 .01
LD-Distributed 12 .44 n.s.
LD-Slow One-trial 17 .28 n.s.
Passage 2

HD-Massed 19 .66 .01
HD-Distributed 18 .50 .05
HD-Slow One-trial 16 .09 n.s.
LDuMassed 10 .42 n.s.
LD-Distributed 12 .82 .01

LD-Slow One-trial 17 .38 ' n.s.

 

 

These correlation coefficients Show that I.Q. and
comprehension are significantly correlated under LD-Massed
(in Passage 1 only). LD-Distributed (in Passage 2 only). HD-
Massed (in Passage 2 only), HD-Distributed (in Passage 2 only),
and LD-Distributed (in Passage 2 only) conditions.

Table 22 presents correlations between verbal reasoning

ability and recall of Passage 1 and Passage 2.

56

Table 22. Correlations between verbal reasoning ability and
recall.

 

 

 

 

 

Group N r P

Passage 1

HD-Massed 19 .32 n.s.
HD-Distributed 18 .28 n.s.
HD-Slow One-trial 16 .76 .01
LD-Massed 10 .65 .05
LD-Distributed 12 .46 n.s.
LD-Slow One-trial 17 .55 .02
Passage 2

HD-Massed 19 .19 n.s.
HD-Distributed 18 .31 n.s.
HD-Slow One-trial 16 .16 n.s.
LD-Massed 10 .20 n.s.
LDinistributed 12 .48 n.s.
LD-Slow One-trial 17 -.04 n.s.

 

 

With the exception of a few significant correlations
between verbal reasoning ability and recall. obtained in
Passage 1. it would appear that there is not a significant
correlation between verbal reasoning ability and recall.

Table 23 presents correlations between verbal reasoning
ability and comprehension.

Table 23. Correlations between verbal reasoning ability and
comprehension.

 

 

 

Group N r P
Passage 1
HD’Massed 19 .27 has.
HD-Distributed 18 .20 n.s.

HID-$101! One-trial 16 -011 11.8 e

57

Table 23 (Continued)

 

 

 

Group N r P

Passage 1

LD-Massed 10 ' .81 .01
LD-Distributed 12 .29 n.s.
LD-Slow One-trial 17 .52 .05
Passage 2

HD-Massed 19 .55 .02
HD-Distributed 18 .50 .05
HD-Slow One-trial 16 .02 n.s.
LD-Massed 10 .28 n.s.
LD-Distributed 12 .59 .05
LD~Slow'0ne-trial 17 .12 n.s.

 

 

Again. as with recall. verbal reasoning ability does
not seem to be significantly correlated with comprehension.
A comprehensive discussion of all the results presented

in Chapter IV is presented in the next chapter.

CHAPTER V
DISCUSSION AND CONCLUSIONS

Analysis of the items from Passage 1 Showed that, as
predicted. recall of high difficulty material under the
distributed condition was significantly better than recall
of high difficulty material under the massed condition.
Further, recall of high difficulty material under the Slow
one-trial condition was significantly better than recall of
high difficulty material under either the distributed condi-
tion or massed condition. Recall of low difficulty material
was Significantly better than recall of high difficulty
material in the massed condition.

Analysis of the comprehension items from Passage 1
Showed. with the exception of the interaction between mode
of presentation and grammatical complexity, that the locus
of comprehension was as predicted. Comprehension of low
difficulty material was significantly better than comprehen-
sion of high difficulty material. under either the
distributed or massed conditions. Comprehension of high
difficulty material under the Slow one-trial condition was
significantly better than comprehension of high difficulty
material under the massed condition.

The presence of interaction between mode of presentation

and grammatical complexity, in recall of Passage 1 indicated,

58

59

as expected, that the differences between the means of high
difficulty material were significantly greater than the
differences between the means of low difficulty material.

And the lack of predicted interaction between mode of presenta-
tion and grammatical complexity in comprehension of Passage 1
was consistent with the theoretical assumption that the
presence of a certain amount of recall does not necessarily
mean that there is a corresponding amount of comprehension.
Thus. although there was a significant interaction between
mode of presentation and grammatical complexity in recall.
there was not a corresponding significant interaction between
mode of presentation and grammatical complexity in
comprehension.

Examination of number of words in the low difficulty
version and the high difficulty version of Passage 1 showed
that the low difficulty version had 354 words and the high
difficulty version had 376 words. There is also a similar
Situation in Passage 2. Consequently. a competing hypothesis
could be advanced that those obtained predicted significant
differences due to level of difficulty were not due to level
of difficulty but due to the fact that there were more words
in the high difficulty version than there were in the low
difficulty version (Savin and Perchonock, 1965). Another
competing hypothesis could also be advanced that the obtained
predicted differences due to level of difficulty were not
entirely due to level of difficulty alone. but also partly
due to the fact that the high difficulty version had more

60

words than the low difficulty version.

However. since the amount of information in the two
versions is the same. and since the difference between the
number of words in the high difficulty version is very small.
it is unlikely that any of the above hypotheses are tenable.
Moreover. length of a prose passage does not seem to be a
Significant factor in how well the passage is comprehended.

Analysis of performance on recall in Passage 2 also
Showed. as expected. that there was a significant main effect
due to difficulty. The recall of low difficulty material
was significantly better than recall of high difficulty
material. under the massed condition and the distributed
condition.

The unexpected lack of significant main effect due to
mode of presentation in recall of Passage 2 was probably due
to the fact that the §§ heard the second passage read to them
on the tape-recorder after they had already listened to the
first passage read at the same Speed. The §§ had, therefore,
practiced under a given mode of presentation and the
generalization of this prior practice may account for the
lack of a main effect due to mode of presentation. This is
consistent with the fact that S5 can be trained to listen and
comprehend Speeded discourse. This interpretation is
supported by some empirical data. For example, in one study
(Orr. Friedman.& Williams, 1965) SE were given a systematic
practice in listening comprehension of prose material tape-

recorded at progressively increasing rates of 325 wpm to

61

475 wpm. The results showed that increasing speed of presenta-
tion in words per minute up to double the normal rate (175 wpm)
did not produce a Significant loss in comprehension.

An analysis of performance on comprehension of Passage
2 also Showed, as expected, that there were Significant main
effects due to levels of difficulty and modes of presentation.
Comprehension of low difficulty material was significantly
better than comprehension of high difficulty material. under
the massed condition and the distributed condition. Comprehen-
sion of high difficulty material under the slow one-trial
condition was Significantly better than comprehension of high
difficulty material under massed condition.

The lack of a significant effect due to mode of presenta-
tion in recall of Passage 2 and the presence of such a
significant effect in comprehension of Passage 2 indicates
that while generalization of prior practice effects led to
comparable performance on recall of factual information. this
phenomenon was not found in comprehension. Again. this
finding is consistent with the theoretical assumption that
comprehension involves more complicated cognitive processes
than recall. Thus, although mode of presentation was not
important in recall of factual information. after prior
practice. it was important in the organization and under-
standing (comprehension) of the material.

Analysis of correlations between recall and comprehension
generally revealed that recall and comprehension of high
difficulty material were not significantly correlated. 0n

62

the other hand, recall and comprehension of low difficulty
material were generally found to be significantly correlated.
Examination of the significant correlation coefficients
showed that only a small percentage of the variance in
comprehension can be accounted for by the variation in recall.

These results seem to suggest that trials or number of
presentations per se is not the major factor facilitating
recall and comprehension of low difficulty and high difficulty
prose material. This disproves the original prediction that
trials or number of presentations would facilitate the
acquisition of factual material in the low difficulty passages.
These results also seem to indicate that difficult trans-
formations require inter-sentence and intra-sentence intervals
long enough so that the transformations can be made. This
is consistent with the original prediction that recall and
comprehension of high difficulty material would be better
under a distributed condition (in which there were 5-Second
pauses between sentences or inter-sentence intervals) and
also under a Slow one-trial condition (in which there were
5-second pauses between sentences and some short pauses
between words in the sentence).

Analysis of correlations between I.Q. and recall
revealed some interesting results. There was a significant
correlation between I.Q. and recall of high difficulty
material under a Slow one-trial condition. There was also
a significant correlation between I.Q. and recall of low

difficulty material under the massed (in Passage 1 only) and

63

distributed conditions. These findings are interpreted to
mean that high I.Q. §§ under the HD-Slow One-trial condition.
LD-Massed condition. and LD-Distributed condition are process-
ing information faster and more efficiently than the low §_s_.
The lack of significant correlations between I.Q. and recall
under the HD-Massed and HD-Distributed conditions means that
the material was so difficult and was presented so fast that
even the high I.Q. §§ were performing as poorly as the low
I.Q. §§. Similarly. the lack of significant correlations
between I.Q. and recall under the LD-Slow One-trial condition
means that the material was so easy and was presented at such
a Slow rate that even the low I.Q. §§ were performing as well
as the high I.Q. §§.

Analyses of correlations between I.Q. and comprehension
generally Showed that I.Q. and comprehension were signi-
ficantly correlated under the LD-Massed. (in Passage 1 only),
HD-Massed (in Passage 2 only). HD-Distributed (in Passage 2
only). and LD-Distributed (in Passage 2 only) conditions.
This is interpreted to mean that high I.Q. §§ were organizing
the acquired information better than low I.Q. Sg. under these
conditions.

Examination of correlations between verbal reasoning ability
and recall generally indicated that verbal reasoning ability
is neither correlated with recall nor comprehension. This
essentially means that verbal reasoning ability is not a good
predictor of how well a.§lwill recall and comprehend a prose

passage.

64

At any rate, the correlations between individual differ-
ence measures and the dependent variables should be regarded
as tentative findings since individual difference measures
were not available for all 120 §§n However, these tentative
findings Should precipitate great interest in the nature of
the interaction of individual differences with modes of
presentation in recall and comprehension of low and high
difficulty materials. An investigation in this area may en-
able one to specify certain educational prescriptions which
permit better learning for some individuals than others.

In general. the results indicated that some of the find-
ings of the sentence studies on grammatical transformations
could be generalized to prose learning.

Number of trials or presentations per se does not seem
to be the major factor facilitating learning of low difficulty
and high difficulty prose materials.

Difficult grammatical transformations require inter-
sentence and intra-sentence intervals long enough to that the
transformations can be made.

On the whole. the results support the grammatical trans-
formational model discussed in Chapter 1.

Any conclusions regarding the relation between the
individual difference measures and the dependent variables
should be regarded as tentative until further information is
obtained. The only conclusion that can be made on the basis
of the available information is that for low I.Q. children

both low and high difficulty prose materials should be

65

presented at a very Slow rate. But for high I.Q. children
slow presentation should be necessary only for high difficulty
prose materials.

Lastly, mode of presentation does not seem to be an
important factor in presenting material to children of high
reasoning ability and low reasoning ability.

CHAPTER VI
SUMMARY

The main objective of this study was to examine the kind
of manipulations which possibly influence recall and comprehen-
sion of prose materials. 7

A 354-word standardized reading passage for grade nine
was selected from "Tests of Academic Progress: Form 1." The
average sentence length of this passage was 16.8 words.
Another standardized reading passage (655 words) was selected
from "Iowa Tests of Basic Skills: Form 3." The average length
of this passage was 13.9 words. These two passages constituted
low difficulty versions.

The following procedures were employed in transforming
these Low'Difficuity Versions (LDV) into High Difficulty
Versions (HDV): Wherever possible, some sentences in the LDV
were changed into passives. noun clauses, embedded versions,
nominalized versions, and then. wherever possible, combining
these transformed versions into long complex sentences.

Three different presentation modes were employed:

1) Both the LDV and HDV were read twice on a tape-recorder,
at a fast rate (Massed condition). 2) The LDV and HDV were
read twice at a fast rate (as in massed condition) but there
were 5-Second pauses at the end of each sentence (Distributed
condition). 3) The LDV and HDV versions were read once at

a very slow rate (Slow One-trial condition). There were,

66

67

thus, six experimental conditions: High Difficulty ~Massed
(HD-Massed), High Difficulty-Distributed (RD-Distributed),
High Difficulty-Slow One-trial (HD-Slow One-trial). Low
Difficulty-Massed (LD-Massed). Low Difficulty-Distributed
(LD-Distributed), and Low Difficulty Slow One-trial (LD-Slow
One-trial).

One hundred and twenty gs were assigned equally to six
experimental conditions.

Some incidental information regarding individual differ-
ence measures (I.Q. and verbal reasoning ability scores) were
available for some of the §§n This incidental information
was useful in the examination of whether or not any of these
individual difference measures is significantly correlated
with any of the dependent variables under any of the six
experimental conditions.

It was hypothesized that:

A. Egggll:

1. Mode of presentation and grammatical
complexity will interact.

2. Recall of low difficulty material
will be better than recall of high
difficulty material, under the massed
condition.

3. Recall of high difficulty material
will be better under the distributed
condition than under the massed

condition.

5.

68

Recall of high difficulty material
will be better under the Slow one-
trial condition.

Recall of low difficulty material

will be better under a massed condition.

B. Comprehension:

1.

2.

3.

5.

Mode of presentation and grammatical
complexity will interact.

Comprehension of low difficulty
material will be better than comprehen-
sion of high difficulty material.

under a massed condition.

Comprehension of high difficulty mater-
ial will be better under a distributed
condition than under a massed
condition.

Comprehension of high difficulty
material will be better under a slow
one-trial condition.

Comprehension of low difficulty mater-
ial wiIl be better under a massed

condition.

The results indicated that number of presentations was

not a Significant factor facilitating learning of prose mater-

ial. Thus, the hypotheses that had predicted better recall

69

and comprehension of low difficulty materials, under a massed
condition. were not confirmed. 0n the whole, all the other
predictions were confirmed. These results were interpreted
as supporting a transformational grammar model.

The analysis of the correlations between individual dif-
ference measures and the dependent variables indicated that
I.Q. and recall were significantly correlated under HD-Slow
one-trial. LD-Massed (except in Passage 2), and LD-Distributed
conditions. There were also significant correlations between
I.Q. and comprehension under LD-Massed (in passage 1 only),
LD-Distributed (in Passage 2 only). HD-Massed (in Passage 2
only). HD-Distributed (in Passage 2 only). and LD-Distributed
(in Passage 2 only) conditions.

With the exception of a few Significant correlations be-
tween verbal reasoning ability and recall, it appeared that
verbal reasoning ability and recall were not Significantly cor-
related. Similarly. verbal reasoning ability did not seem to
be significantly correlated with comprehension.

In conclusion. it can be said that number of presentations
is not a significant factor facilitating learning of prose
material. Recall and comprehension of prose material contain-
ing complex grammatical constructions tend to be significantly
poorer. However. there was a significant interaction between
mode of presentation and level of grammatical complexity. The
slow one-trial mode greatly facilitated the recall and comprehen-
sion of grammatically complex prose material. These results

suggest that complex grammatical transformations require

70

transformational time to reduce them to their kernel sentences.
Thus, the greater intra- and inter-sentence intervals in the
Slow one-trial condition allow §§_the necessary transformation

time.

APPENDIX A

PRETEST PASSAGE

Rio de Janeiro often is called Rio. It was once the
capital and is still the largest city in Brazil. It lies on
a bay near the Atlantic Ocean. Its outstanding landmarks
are a 1.200-foot peak at the entrance to the bay and a huge
concrete statue of Christ atop 2.300-foot Corcovado Mountain.

The green islands in its bay. its white beaches, its blue
harbor. and its gray hills provide Rio with a most beautiful
setting. Among manpmade additions to its charm are its mosaic
sidewalks, its plazas, and its tree-lined boulevards.

R10 is located in Southeast Brazil. It extends from the
bay. along the coast, and into the valleys of nearby mountains.
Granite hills, many of which have been tunneled. divide the
city into sections. Natives as well as tourists are attracted
to its botanical gardens, its opera house. its national
library. its lavishly ornate public buildings and churches.
One of the most awe-inspiring of its sights is its Municipal
Stadium packed with 150,000 spectators at a musical event,
sports contest, or political rally.

The Portugues colonized Brazil and they ruled it for
centuries. They gave Brazil its Catholic religion and its
Imational language. The rights of Portugal to Brazil became.

established through a treaty with Spain in 1494, and Brazil
did not achieve an independent government until 1822. Even
With independence. Brazil's government retained a Portuguese
flavor through a monarchy that remained in power until 1889.
Brazil became a republic under a constitution in 1891.

71

72

Passage 1

Low Difficulty Version

Man's paintings give us a record of his culture and his
problems from a time long before he acquired the ability to
prepare a written history. The walls of European caves have
paintings that have survived from the early Stone Age.

Although many of primitive man's drawings consisted of
"stick figures," some of the best cave paintings Show form
and movement that are amazingly accurate and realistic.

The paintings of ancient peoples tell us of their prob-
lems and their heroes. From their art we can learn about their
government and their religion. For example. we know that the
Egyptians had powerful and glorified rulers and their religion
included a belief that there was life after death.

From their early, crude recordings of specific events,
theEgyptians developed a complex set of written symbols,
hieroglyphics.

Much of our present knowledge concerning the early
civilization of China and Greece comes from information
recorded in various art forms. From pictures, we know of the
Greek gods and of the Greeks' reverence for athletic prowess.

The Babylonians early developed the art of making
pictures with glazed tile. These extremely colorful and
permanent pictures contained symbolic designs and patterns.
As a development from tile art came mosaic pictures formed

from bits of colored glass and marble. Many of these

73

featuring religious subjects, are preserved on the walls of
early churches.

During the period of medieval art, from approximately
500 A.D. to 1500, frescoes and the "illumination" of manu-
scripts became established as mature art forms. Fresco
painting is done on.Fresh (hence the name) plaster. The
design is first sketched and then the pigment is combined with
the plaster. Fresco painting is well suited to large surfaces.
Illumination was developed by the monks and resulted in
beautiful and intricate letters and pictures on their scrolls.
Modern day artists find difficulty in improving upon or even
matching the quality of the art of this period.

Shortly after 1400 A.D.. artists began to paint with 011.
first on wood and then on canvas. The use of slow-drying oil
allowed for great detail and the many possible colors

encouraged brilliance, shading, and depth.

74

Passage 1
High Difficulty Version

A record of man's culture and his problems from a long
time before he acquired the ability to prepare a written
history is given us by his paintings. Paintings on the walls
of European caves have survived from the early Stone Age.

Form and movement that are amazingly accurate and
realistic are shown by some of the best cave paintings, al-
though many of primitive man's drawings consisted of stick
figures.

By the paintings and art of Ancient pe0p1es we are told
of their problems and heroes, their government and religion.
That the Egyptians had powerful and glorified rulers, for
example, and that included in their religion was a belief that
there was life after death. we know.

A complex set of written symbols, hieroglyphics. was
developed by the Egyptians from early. crude recordings of
specific events that they had made.

Coming from information recorded in various art forms.
is much of our present knowledge concerning the early civiliza-
tion of China and Greece, and the Greek gods and the Greeks'
reverence for athletic prowess are known by us. from pictures.

The art of making pictures with glazed tile was early
developed by the Babylonians. and symbolic designs and patterns
were contained in these extremely colorful and permanent
pictures. Mosaic pictures that were a development from tile

art were formed from bits of colored glass and marble, and

75

many of these, featuring religious subjects, are preserved
on the walls of early churches.

The establishment of frescoes and the "illumination" of
manuscripts as mature art forms took place from approximately
500 A.D. to 1500, during the period of medieval art. On
fresh (hence the name) plaster. is done Fresco painting which
is well suited to large surfaces, and combination of the pig-
ment with the plaster takes place after the design is first
sketched. Illumination which was developed by the monks
resulted in beautiful and intricate letters on their scrolls
and difficulty is found by modern-day artists in improving
upon or even matching the quality of the art of this period.

Painting with oil, first on wood and then on canvas,
was begun by artists Shortly after 1400 A.D.. great detail
being allowed through use of Slow-drying oil. and brilliance,
shading and depth were encouraged by many possible colors.

76

Passage 2

Low Difficulty Version

The Indians called him "Yellow Hair." Most people of
his day considered George Armstrong Custer a dashing figure.
The youthful cavalry officer was impetuous and daring in
action. romantic in appearance. He fancied embroidered buck-
skins. leather gauntlets. and big-rimmed hats. His curly.
red-gold hair swept his shoulders. He has been called
conceited. headstrong, and foolish - and also one of the
finest horsemen the Army ever had.

Custer was born in Ohio in 1839. From boyhood he wanted
to be a soldier. and yet he graduated from West Point at the
foot of his class. His teachers described him as intelligent
but careless. A few days after graduation. he was ordered
to duty in the Civil War. During the next two years his
record for bravery was so outstanding that by 1863 he had
risen to the wartime rank of brigadier general, the youngest
man of that rank in the Army at the time. His youth. his
fearlessness, and his conspicuous dress made him a great
favorite.

When the war ended. Custer was given a lower rank and
sent West to fight the Indians. Bored and impatient. he was
often in trouble with his superiors. After fighting the
Cheyennes in Kansas, he was sent north to protect the build-
inf of the Northern Pacific Railroad. In 1874 he helped a
group of scientists explore the Black Hills of South Dakota.

77

the territory of the fierce Sioux tribes.

When gold was discovered there, the Sioux grew restless.
They did not like miners poking around their reservations.
Furthermore. the railroad was creeping toward their territory.
White settlers were moving onto land over which the Indians
had hunted for generations. The Sioux foresaw the end of the
buffalo herds on which they depended for meat and clothing.
Embittered young warriors quietly slipped away from the
reservations and joined war camps. There were sudden attacks
on settlers. Alarmed, the government in Washington ordered
all Indians to return to their reservations. but this order
was contrary to certain treaties that allowed the Sioux to
hunt on the northern plains. Two of the strongest chiefs,
Sitting Bull and Crazy Horse. refused to obey and prepared
to fight.

Early in the Spring of 1876. General Alfred.Terry was
sent out from.Bismarck, North Dakota. to bring the two chiefs
in. His troops included the Seventh Cavalry Regiment, which
Custer commanded. When Terry's men reached Montana. advance
scouts located Sitting Bull's camp on the Little Bighorn River.
Custer and 600 men were sent ahead to bar the chief's escape
until the rest of the troops could join them.

At the bank of the river. Custer pondered his next move.
Though clouds of smoke and dust rose beyond the hills on the
other side, he could not judge the size of the Indian camp.
Nor did he know whether the Indians guessed how small his own

forces were. Should he wait for General Terry. or should he

78

attack at once? If he waited for General Terry. the Indians
might break camp and flee, or they might attack him. Perhaps,
Custer wanted the glory of capturing Sitting Bull himself.

At any rate. he decided not to wait for Terry.

He dispatched part of his troops to attack the Sioux
camps to the South. The remaining 265 he led across the
Little Bighorn toward the center of the Sioux line.
Immediately. between 2500 and 4000 whooping Indians poured
out from the hills. Somehow Custer's men managed to fight
their way to a higher hill. There a thousand Cheyennes
suddenly appeared, and it was all over. Only one member of
Custer's band, a Crow scout. escaped. "Yellow Hair" and two
younger brothers died with the rest. All were scalped except
the commander.

Not until General Terry arrived was the extent of the
disaster known. The news stunned the nation.) Custer became
a popular but controversial hero. The battlefield near
ZHardin. Montana. where he made his "last stand.“ is now a

national monument.

79

Passage 2

High Difficulty Version

Impetuous and daring in action, and romantic in appear-
ance. the youthful cavalry officer. George Armstrong Custer,
whose shoulders were swept by his curly red-gold hair. was
called "Yellow Hair" by the Indians, and a dashing figure by
most people of his day. Embroidered buckskins, leather gaunt-
lets. and big-rimmed hats were fancied by him, and although
he has been called conceited, headstrong and foolish. he was
one of the finest horsemen ever had by the Army.

Born in Ohio in 1839, and described as intelligent but
careless by his teachers, Custer wanted to be a soldier from
boyhood even though he graduated from West Point at the foot
of his class. That his record for bravery was so outstanding
was shown by the fact that when he was ordered to duty in the
Civil War a few days after graduation, his performance during
the next two years led to his rising to the wartime rank of
brigadier general, the youngest man of that rank in the Army
at the time, by 1863. His being a great favorite was made
by his youth, his fearlessness, and his conspicuous dress.

Bored and impatient, Custer was often in trouble with
his superiors after he had been given a lower rank and was
sent west to fight the Indians when the war ended. A group
of scientists was helped by him to explore the Black Hills
of South Dakota, the territory of the fierce Sioux tribes,

in 1874, after the Cheyennes in Kansas had been fought by him

80

and he having been sent north to protect the building of the
Northern Pacific Railroad.

The Sioux, who did not like miners poking around their
reservations. and. furthermore, whose territory the railroad
was moving toward, grew restless when gold was discovered
there. That the Sioux foresaw the end of the buffalo herds
on which they had depended for meat and clothing was the
result of the moving onto land over which the Indians had
hunted for generations, by White settlers. Sudden attacks on
settlers were made by embittered young warriors who quietly
slipped away from the reServations and joined war camps. Two
of the strongest chiefs, Sitting Bull and Crazy Horse, who
refused to obey an order by the alarmed government in
Washington that all Indians return to their reservations in
that this order was contrary to certain treaties that allowed
the Sioux to hunt on the northern plains, prepared to fight.

General Alfred Terry. whose troops included the Seventh
Cavalry Regiment. which Custer commanded, was sent out from
Bismarck. North Dakota, to bring the two chiefs in, early in
the spring of 1876. Sitting Bull's camp on the Little Bighorn
River was located by advance scouts when Montana was reached
by Terry's men. and the chief's escape was barred by Custer
and 600 men who had been sent ahead until the rest of the
troops could join them.

Pondering his next move. whether to wait for General
Terry or attack at once. at the bank of the river. and though

clouds of smoke and dust rose beyond the hills on the other

81

side, the size of the Indian camp could not be judged by him.
nor was it known by him whether it had been guessed by the
Indians how small his own forces were. Although Custer may
have wanted the glory of capturing Sitting Bull himself, per-
haps, the fact is he decided not to wait for Terry because

if he had waited. the camp might have been broken by the
Indians and they would have fled or he might have been attacked
by them.

Part of his tr00ps were dispatched by him to attack the
Sioux camps to the south. and the remaining 265 were led by
him across the Little Bighorn toward the center of the Sioux
line. Somehow Custer's men managed to fight their way to a
higher hill when between 2500 and 4000 wildly whooping Indians
poured out from the hills. immediately. Only one member of
Custer's band. a Crow scout, escaped when it was all over as
a result of a sudden appearance of a thousand Cheyennes.
Dying with the rest were "Yellow Hair" and two younger brothers,
and all. except the commander. were scalped.

Not until General Terry arrived was the extent of the
disaster made known by him. The nation was stunned by the
news. That Custer became a popular but controversial hero
is evident from the fact that the battlefield near Hardin.
Montana, where he made his "last stand" is now a national

monument.

l.

3.

APPENDIX B

Questions for Pretest

What is the national language of Brazil?
1) Spanish I
2) Portuguese

3) Brazilian

4) English

5) The selection gives no clue

Which of the following divides Rio into sections?
1) Bays

2) Valleys

3) Tunnels

4) Hills

5) Tree-lined boulevards

What are the principal attractions for tourists of Rio?
1) Its musical events

2) Its sports contests

3) Its religious shrines

4) Its climate and beautiful beaches

5) Its gardens and public buildings

According to the selection. which of the following terms
best describes Brazil's churches?

1) Lavish

2) Huge

3) Catholic

4) Charming

5) Ornate 82

50

83

The author of this selection refers to Rio's gray hills.
From clues given in the selection, what factor seems
most responsible for this appearance?

1)
2)
3)
4)
5)

Lack of moisture
Hot climate
Exposed rock
Low-hanging clouds

Industrial smoke

According to the selection, what are the most outstanding
landmarks in Rio?

1)
2)
3)
4)
5)

A small mountain and a statue

Its magnificent old churches

Its stadium and other public buildings
Its mosaic Sidewalks and its parks

Its beautiful bays and beaches

What form of government does Brazil have today?

1)
2)
3)
4)
5)

Limited monarchy
Monarchy
Republic
Democracy

The selection does not discuss today's government

What is Ric's relationship to Brazil?

1)
2)
3)
4)
5)

Capital city
Largest city
Most beautiful city
An independent city

Most religious city

2.

3.

5.

1)
2)
3)
4)
5)

1)
2)
3)
4)
5)

1)
2)
3)
4)
5)

1)
2)
3)
4)
5)

1)
2)
3)

84

Response Sheet for Pretest

Spanish
Portuguese
Brazilian

English

The selection gives no clue

Bays

Valleys

Tunnels

Hills

Tree-lined boulevards

Its musical events

Its sports contests

Its religious shrines

Its climate and beautiful beaches

Its gardens and public buildings

Lavish
Huge
Catholic
Charming

Ornate

Lack of moisture
Hot climate

Exposed rock

7.

8.

4)
5)

1)
2)
3)
4)

1)
2)
3)
4)
5)

1)
2)
3)
4)
5)

85

Low-hanging clouds

Industrial smoke

A small mountain and a statue
Its magnificent old churches.
Its stadium and other public buildings

Its mosaic sidewalks and its parks

Limited monarchy
Monarchy
Republic
Democracy

The selection doesnot discuss today's government

Capital city
Largest city
Most beautiful city
An independent city

Most religious city

1.

2.

3.

86

Questions for Passage 1

When were Frescoes and the illumination of manuscripts
become established as mature art forms?

1) During the Stone Age

2) During the period of Medieval art

3) After 1500 A.D.

4) Before man acquired the ability to prepare a written
history

5) None of these

According to the selection, what is manuscript "illumina-

tion"?

1) Explaining what the writer means

2) Furnishing sufficient light to read

3) Providing charts and graphs

4) Decorating the manuscript with color and design

5) Making the manuscript easier to read

From the selection it can be concluded that Egyptian
rulers were

1)
2)
3)
4)
5)

Immoral
Autocratic
Religious
Democratic

None of these

Many of primitive’man's drawings consisted of

1)
2)
3)
4)

Complex symbols
Scrolls

Stick figures
Symbolic designs

5.

7.

5)

The

87

None of these

walls of European caves have paintings that have

survived from

1)
2)
3)
4)
5)

The medieval period
The Early Stone Age
The mosaic period
Shortly after 1400 A.D.

None of these

Frescoes and illumination of manuscripts became esta-
blished as a mature art form during

1)
2)
3)
4)
5)

The Stone Age

The period of medieval art
The Babylonian period

The mosaic period

None of these

Mosaic pictures were a development from

1)
2)
3)
4)
5)

The
1)
2)
3)
4)
5)

Medieval art
Fresco painting
Tile art )
Stick figures

None of these

illumination of manuscripts was developed by
The Egyptians

The Babylonians

The nuns

The Chinese

The monks

9.

10.

11.

12.

88

0f the materials discussed in the selection. which of
the following. if any, was most recently intorduced as
a work surface?

1)
2)
3)
4)
5)

Scrolls
Canvas
Wood
Stone

The selection gives no clue

According to the selection. what is the principle advant-
age of water color over other art mediums?

1)
2)
3)
4)
5)

{A wider variety of colors is possible
Cheaper materials are available

Larger surfaces can be covered more easily
All the above are advantages

The selection gives no clue

According to the selection. to which of the following.
if any, can the origin of mosaic art be traced?

1)
2)
3)
4)
5)

Primitive art
Egyptian
Babylonian
Medieval

The selection gives no clue

What does the term "fresco" mean?

1)
2)
3)
4)
5)

Plaster
Fresh
Wet
Colored

The selection gives no clue

89

13. Who among the following develOped the art of making
pictures with glazed tile?

1) The Egyptians
2) The Chinese
3) The Babylonians
4) The Greeks
5) None of these
14. According to the selection, which period, if any.
produced an art form most difficult for modern-day
artists to duplicate?
l) Primitive
2) Babylonian
3) Egyptian
4) Medieval

5) The selection gives no clue

15. On which of the following is Fresco painting done?
1) On the wall
2) 0n fresh plaster
3) On stones
4) On.wood
5) On canvas
16. The walls of European caves have paintings that have
survived from
1) The mosaic period
2) The Medieval Age
3) Early Stone Age
4) Babylonian period
5) None of these

17.

18.

19.

20.

90

According to the selection. for what were hieroglyphics

used?

1) Decorating walls

2) Recording religious events
3) Recording information

4) Illuminating manuscripts
5) The selection does not say

According to the selection. what art form is well suited
to covering the wall of a large room?

1)
2)
3)
4)
5)

Mosaic

Oil on canvas
Water color
Fresco

The selection gives no clue

A complex set of written symbols developed by the
Eygptians were known as

1)
2)
3)
4)
5)

Braille
Microscopics
Hieroglyphics
Cuineform

None of these

According to the selection. what can be said about the
Ancient man?

1)
2)
3)

The Ancient man had no culture before a written history
The Ancient man lived in European caves

The government and religion of the Ancient man can
be learned from the paintings of ancient peoples

91

4) The Ancient man was a cannibal

5) None of these

l.

2.

3.

1)
2)
3)
4)

5)

1)
2)
3)
4)
5)

1)
2)
3)
4)
5)

1)
2)
3)
4)
5)

1)
2)

Response Sheet for Passage 1

During the Stone Age
During the period of Medieval art
After 1500 A.D.

Before men acquired the ability to prepare a
written history

None of these

Explaining what the writer means

Furnishing sufficient light to read

Providing charts and graphs

Decorating the manuscripts with color and design
Making the manuscripts easier to read

Immoral
Autocratic
Religious
Democratic

None of these

Complex symbols
Scrolls

Stick figures
Symbolic designs

None of these

The medieval period
The Early Stone Age

92

7.

9.

3)
4)

5)

1)
2)
3)
it)
5)

1)
2)
3)
1+)
5)

1)
2)
3)
4)
5)

1)
2)
3)
4)
5)

93

The mosaic period
Shortly after lhoo A.D.

None of these

The Stone Age

The period of medieval art
The Babylonian period

The mosaic period

None of these

Medieval art
Fresco painting
Tile art

Stick figures

None of these

The Egyptians
The Babylonians
The nuns

The Chinese
The monks

Scrolls
Canvas
Wood
Stone

The selection gives no clue

10.

ll.

12.

13.

11+.

1)
2)
3)
1+)
5)

1)
2)
3)
4)
5)

1)
2)
3)
’4)
5)

1)
2)
3)
1+)
5)

1)
2)
3)
1+)
5)

94

A wider variety of colors is possible
Cheaper materials are available

Larger surfaces can be covered more easily
All the above are advantages‘

The selection gives no clue

Primitive art
Egyptian
Babylonian
Medieval

The selection gives no clue

Plaster
Fresh
Wet
Colored

The selection gives no clue

The Egyptians
The Chinese
The Babylonians
The Greeks

None of these

Primitive
Babylonian
Egyptian
Medieval

The selection gives no clue

15c

16.

17.

18.

19.

1)
2)
3)
1*)
5)

1)
2)
3)
4)
5)

1)
2)
3)
4)
5)

l)
2)
3)
4)
5)

1)
2)
3)
1+)
5)

95

On the wall

On fresh plaster
On stones

On wood

On canvas

The mosaic period
The medieval age
Babylonian period
Early Stone Age

None of these

Decorating walls
Recording religious events
Recording information
Illuminating manuscripts

The selection does not say

Mosaic

Oil on canvas
Water color
Fresco

The selection gives no clue

Braille
Microscopics
Hieroglyphics
Cuineform

None of these

20.

1)

2)
3)

1*)
5)

96

The Ancient man had no culture before a written
history

The Ancient man lived in European caves

The government and religion of the Ancient man can
be learned from the paintings of ancient peoples

The Ancient man was a cannibal

None of these

1.

3.

Questions for Passage 2
Which of the following would make the best title for
this selection?
1) Custer the Great
2) The Yellow Hair
3) The military career of George Armstrong Custer
u) The death of Custer
What effect did Custer's Civil War experience have on
his personality?
1) It made him cautious
2) It caused him to have a very high opinion of himself
3) It killed his ambition
n) It made him disagreeable and unfriendly
In terms of the size of the opposing forces, what were
the odds against Custer and his men?
1) About 2 to l
2) About 15 to l
3) Amost 100 to 1
A) Better than 1000 to 1

In which state did Custer make his "last stand"?
1) Kansas

2) North Dakota

3) Montana

4) South Dakota

97

5.

9.

98

Why did the Indians resent the coming of the white men?

1)
2)
3)

They wanted to mine the gold for their own use
They were afraid of railroad trains

They feared the loss of their food and clothing supply

What respect did the Indians show for the held Custer?

1)
2)
3)
1+)

They took him prisoner instead of killing him
They gave him a military funeral
They built a monument in his honor

They did not scalp him

What is the Little Bighorn?

1)
2)

3)
4)

An Indian reservation

A river

One of the Black Hills

A pass between two high hills

Which phase of Custer's career is treated in the greatest

1)
2)
3)
1*)

'detail in this article?

His training at West Point
His Civil War experience
His postwar assignments in Kansas and South Dakota

The part played in the campaign against the Sioux

How many men died with Custer?

1)
2)
3)
4)

200
261)
600

No one knows for sure

10.

11.

12.

l3.

14.

99

What error in Judgement did Custer make at the Little
Bighorn?

1) He underestimated the size of the enemy forces
2) He delayed his attack too long
3) He attacked across a river

4) He tried to fight from a hill

Who among Custer's forces escaped the massacre?
l) Custer himself

2) A younger brother of Custer

3) An Indian scout

4) No one escaped

Custer commanded a regiment of
1) Foot soldiers
2) Horsemen

3) Engineers

4) Indian scouts

Why did General Terry send Custer to the Little Bighorn?

1) To capture Sitting Bull
2) To chase the Indians back to their reservations

3) To round up some Indians who could be trained as
scouts

4) To hold the Indians in Sitting Bull's camp

Why were the Indians ordered to return to their reservations?

1) To save the buffalo herds from being wiped out

2) To keep them away from the newly discovered gold

15.

16.

17.

18.

a 100

3) Because the treaties they had signed required them
to stay on their reservations

4) To help prevent further attacks on white settlers
What attitude to Custer does the author take in this
article?

1) He believes Custer was a selfish glory hunter, and
tries to prove it

2) He believes Custer was not to blame for the disaster
at the Little Bighorn

3) He believes Custer was one of the finest officers
in the Army

4) He tries to give all the facts and let the reader
decide

Where did Custer receive his formal military training?

1) On the frontier

2) On Civil War battles

3) At the U.S. Military Academy

4) At a military school in Ohio

Why was Custer unhappy in the years following the Civil

War?

1) He missed the excitement and honor that had been
his during the war

2) He was disappointed because the South did not win
the war

3) He was heavily in debt because his army pay was too
low for him to meet his bills

4) He did not like being moved around the country

Which of these things happened last?
1) Custer was promoted to the rank of general

2) Gold was discovered on the Black Hills

101

3) Custer helped protect the building of the Northern
Paolfic

4) Treaties were signed which allowed the Sioux to hunt
on the plains

19. How old was Custer when he was made a brigadier general?

1) 24
2) Z?
3) 30
4) 33

20. The Indians called George Armstrong Custer
l) Blond Hair
2) Red Hair
3) Black Prince
4) Yellow Hair
5) None of these

2.

3.

5.

1)
2)
3)
4)

1)
2)
3)
4)

1)
2)
3)
4)

1)
2)
3)
4)

1)
2)
3)

1)
2)
3)
4)

Response Sheet for Passage 2

Custer the Great

The Yellow Hair

The military career of George Armstrong Custer
The death of Custer

It made him cautious
It caused him to have a very high opinion of himself
It killed his ambition

It made him disagreeable and unfriendly

About 2 to l

About 15 to 1

Almost 100 to 1
Better than 1000 to 1

Kansas
North Dakota
Montana

South Dakota

They wanted to mine the gold for their own use
They feared railroads
They feared the loss of their food and clothing supply

They took him prisoner instead of killing him
They gave him a military funeral
They built a monument in his honor

They did not scalp him

102

7.

8.

9.

10.

11.

12.

1)
2)

3)
4)

1)
2)
3)
4)

1)
2)
3)
4)

1)
2)

3)
4)

1)
2)
3)
4)

1)
2)
3)
4)

103

An Indian reservation

A river

One of the Black Hills

A pass between two high hills

His training at West Point

His Civil War experience

His postwar assignments in Kansas and South Dakota
The part played in the campaign against the Sioux

200
264
600

No one knows for sure

He underestimated the size of the enemy forces
He delayed his attack too long
He attacked across a river

He tried to fight from a hill

Custer himself
A younger brother
An Indian scout

No one escaped

_Foot soldiers
Horsemen
Engineers

Indian scouts

13.

14.

15.

16.

17.

1)
2)

3)
4)

1)
2)
3)

4)

1)

2)

3)

4)

1)
2)

3)
4)

1)

2)

3)

4)

104

To capture Sitting Bull
To chase the Indians back to their reservations
To round up some Indians who could be trained as scouts

To hold the Indians in.Sitting Bull's camp

To save the buffalo herds from being wiped out
To keep them away from the newly discovered gold

Because the treaties they had signed required them
to stay on their reservations

To help prevent further attacks on white settlers
He believes Custer was a selfish glory hunter, and
tries to prove it

He believes Custer was not to blame for the disaster
at the Little Bighorn

He believes Custer was one of the finest officers
in the army

He tries to give all the facts and let the reader
decide .

On the frontier

0n Civil War battles

At the U.S. Military Academy

At a military school in Ohio

He missed the excitement and honor that had been his

during the war

He was disappointed because the South did not win
the war

He was heavily in debt because his army pay was too
low for him to meet his bills

He did not like being moved around the country

18.

19.

20.

1)
2)
3)

4)

1)
2)
3)
4)

1)
2)
3)
4)
5)

105

Custer was promoted to the rank of general
Gold was discovered in the Black Hills

Custer helped protect the building of the Northern
Pacific

Treaties were signed which allowed the Sioux to hunt
on the plains

24

27

30

33

Blond Hair
Red Hair
Black Prince
Yellow Hair

None of these

APPENDIX C

RAN BETA

High Difficulty - Massed Group

 

Passage 2

Passage 1

 

Comprehension Recall

 

Comprehension

Recall

Pretest

8#

 

10

11

12

13
l4
15
l6
17
18

19
20

106

107

High Difficulty - Distributed Group

 

Passage 2

Passage 1

 

 

Comprehension Recall Comprehension

Recall

Pretest

3#

 

10

11

12

13
l4
15
16
17
18
19
20

108

High Difficulty — Slow One-trial Group

 

Passage 2

Passage 1

 

 

Recall

Pretest

3#

Comprehension Recall Comprehension

 

10

11

12

13
14
15
16
17
18
19
20

109

Low Difficulty - Massed Group

 

Passage 2

Passage 1

 

 

Recall Comprehension

Pretest

3#

Recall Comprehension

 

10

11

12

13
l4
15
16
17
18

10

10

10

19
20

110

Low Difficulty - Distributed Group

 

Passage 2

Passage 1

 

 

Recall Comprehension

Pretest

3#

Recall Comprehension

 

10

11

12

13
14
15
16
17
18
19
20

111

Low Difficulty - Slow one-trial Group

 

Passage 2

Passage 1

 

 

Recall Comprehension

S # Pretest

Recall Comprehension

 

10

10

11

12

13
14
15
16
17

11

18
19
20

APPENDIX D
INDIVIDUAL DIFFERENCE MEASURES

High Difficulty - Massed Group

 

 

 

Passage 1 Passage 2 I.Q. Verbal
Reasoning
Recall Comprehension Recall Comprehension Ability

4 0 3 2 107 21
3 4 4 5 114 26
5 2 2 4 102 19
4 4 3 1 88 15
l 2 5 2 94 24
7 4 6 7 124 38
2 3 4 2 96 14
2 6 3 6 126 41
3 l 6 8 108 25
3 l 5 5 100 14
1 3 3 4 92 19
4 3 4 6 126 40
4 4 1 6 99 22
5 6 6 4 114 40
3 6 6 4 106 16
3 3 6 4 93 9
3 3 6 3 101 19
4 1 8 6 121 36
4 3 4 3 98 12

112

113

High Difficulty - Distributed Group

 

 

 

Passage 1 Passage 2 I.Q. Verbal
Reasoning
Recall Comprehension Recall Comprehension Ability
2 3 2 3 94 16
5 2 4 4 120 31
5 1 5 3 104 16
3 5 4 5 124 28
6 3 7 8 120 24
5 6 5 5 100 18
3 6 5 4 97 14
u 7 8 4 124 37
3 2 5 6 103 14
2 4 5 5 125 37
4 5 6 6 118 40
7 4 5 6 120 46
5 4 4 7 124 45
7 5 8 5 111 28
u u 3 2 108 19
6 3 6 6 122 36
5 6 5 3 110 24
6 3 3 1 106 19

114

High Difficulty - Slow One-trial Group

 

 

 

Passage 1 Passage 2 I.Q. Verbal

Reasoning

Recall Comprehension Recall Comprehension Ability
1 3 6 4 114 24
6 7 5 6 114 38
3 3 4 7 107 30
7 4 4 6 128 38
6 6 6 6 123 24
9 7 8 5 144 41
6 4 5 4 122 36
5 5 6 5 108 23
8 4 5 5 127 39
3 7 4 4 106 18
9 5 9 7 134 41
8 5 7 6 132 41
9 7 3 6 118 29
6 7 6 7 117 27
7 5 5 7 128 36
2 4 8 8 116 17

115

Low Difficulty - Massed Group

 

 

 

Passage 1 Passage 2 I.Q. Verbal
Reasoning
Recall Comprehension Recall Comprehension Ability
6 6 3 5 127 35
8 9 5 7 120 35
6 6 5 5 108 33
5 5 6 3 116 28
6 3 5 7 103 19
5 4 6 3 116 32
5 3 8 4 103 21
9 8 6 8 125 29
10 10 10 8 134 44
9 7 6 6 124 39

116

Low Difficulty - Distributed Group

 

 

 

Passage 1 Passage 2 I.Q. Verbal
Reasoning
Recall Comprehension Recall Comprehension Ability
4 5 6 5 100 20
6 5 8 7 113 21
3 5 7 7 112 17
6 9 8 7 112 28
6 5 6 6 108 13
4 4 3 3 82 11
8 4 5 8 112 28
5 4 5 6 106 26
4 2 7 5 109 26
8 9 9 7 122 27
5 4 3 4 98 19
7 4 5 5 112 20

117

Low Difficulty_- Slow One-trial Group

 

 

 

Passage 1 Passagpgz I.Q. Verbal
A Reasoning
Recall Comprehension Recall Comprehension Ability
7 10 5 5 105 34
8 5 6 9 122 29
7 6 5 7 114 26
3 2 7 7 119 29
8 9 6 6 122 33
6 6 5 6 105 16
8 5 6 8 112 28
7 7 8 8 111 36
7 6 4 4 118 28
4 2 8 8 110 18
7 5 6 8 114 17
6 3 4 3 90 16
6 3 5 3 115 30
3 6 6 5 90 9
7 ll 6 6 128 36
5 4 7 6 100 17
6 4 8 6 113 17

BIBLIOGRAPHY

Bruning‘J.L. and Kintz, B.L. Computational Handbook of
Statistics. Scott. Foresman and Company. 1968.

Bunch. M.E. and Magsdick, W.K. The-retention in rats of an
incompletely learned maze solution for short
intervals of time. J. comp Psychol..*l933. 16,
385-409.

Cofer, C.N. A comparison of logical and verbatim learning
of prose passages of different lengths. Amer. J.
Psychol.. 1941, 54. 1-20.

Chomsky, N. Language and the Mind. Psychology Today,
February, 1968, 1. Number 9.

Coleman, E.B. Improving comprehensibility by shortening
sentences. J. appl. Psychol.. 1962, 46. 131-134.

Coleman, E.B. and Blumenfeld, P.J. Cloze Scores of nominaliza-
tions and their grammatical transformations using
active verbs. Psych. Rep., 1963. 13, 651-654.

Coleman, E.B. The comprehensibility of several grammatical
transformations. J. appl. Psychol.. 1964, 48,
NO. 3. 186-190.

Coleman, E.B. Learning of prose written in four grammatical

transformations. J. appl. Psychol.. 1965. 49.
332-341.

Deese, J. The Ps cholo of Learni . McGraw—Hill Book Co.
Inc.: 1958.

Duncan, C.P. The effect of unequal amounts of practice on
motor learning before and after rest. J. exp. Psychol..
1951. “'2. 257-26“.

Edwards, A.L. Statistical Methods for the Behavioral Sciences.
Holt. Rinehart and Winston. 1962.

Fairbanks, G., Guttman, N. and Myron. M. Effects of time
compression upon the compression of connected Speech.
J. Speech Hear. Dis., 1957, 22. 10-19.

Flesch, R.F. The Art of Plain Talk. New York: Harper. 1946.

118

119

Foulke, E.. Amster, C.H.. Nolan. C.Y. and Bixler. R.H. The
comprehension of rapid speech by the blind.

Foulke, E. Proceedings of the Louisville Conference on time
compressed speech (edit.) Center for Rate Controlled
Recordings. Louisville, Kentucky 40208, 1966.

Garvey, W.D. The intelligibility of speeded Speech. J. exp.
Psychol.. 1953, 45. Number 2, 102-108.

Gough. P.B. Grammatical transformations and speed of under-
standing. J. verb. Learn. vepb. Behav.. 1965, 4,
107-111.

Hall. J.F. The Psychology of Learnigg. Lippincott Co.: 1966.

Hays, W.L. Statistics for P8 cholpgists. Holt. Rinehart and
Winston: 1963.

Holland. G.A. Simple trial-and-error learning massed and
distributed trials with habits of unequal initial

strength. J. comp. physiol. Psychol.. 1953, 46,
90-9“.

 

Jesperson. 0. TheBhilospphyfiof Grammar. Holt and Co.:1924.

Keppel, G. Facilitation in short- and long-term retention
of paired associates following distributed practice
in learning. J. verb. Learn. vepp. Behav., 1964,
_3. 91-111.

Kientzle. M.J. Properties of learning curves under varied‘

distribution of practice. J. exp. Psychol.. 1946
36, 187-211. '

Lindquist, E.F. Desi and Anal sis of Experiments in
P8 cholo an? Education. Houghton Miffiin Company.
BostonzI956.

 

Large, I. Influence of regularly interpolated time intervals
upon subsequent learning. Teach. 091;. Cont. Educ..
1930. No. 438.

Martin, C.J. and Alonso. L. Comprehension of Full Length and
Telegraphic Materials among Blind Children.
Educational Research Series, Number 42, December.

1967.

Mehler. J. Some effects of grammatical transformations on
the recall of English sentences. J. verb. Learn.
verb. Behav., 1963, 2. 346-351.

120

Miller. G.A. Some psychological studies of grammar. Amer.

Orr. D.B.. Friedman, H.L. and Williams, J.C.C. Trainabillty
of listening comprehension of speeded discourse.
J. educ. Psychol., 1965, 56, 148-156.

Savin. H.B. and Perchonock. E. Grammatical structure and the
immediate recall of English sentences. J. verb.
Learn. verb. Behav.. 1965. 4. 348-353.

Slamecka. N.J. Studies of retention of connected discourse.
Amer. J. PSIOhO]... 1959. 72. 409-416.

Slamecka, N.J. Retroactive inhibition of connected discourse
as a function of practice level. J. exp. Psychol..

Slobin, D.I. Grammatical transformations and sentence
comprehension in childhood and adulthood. J. verb.
Learn. verb. Behav.. 1966. 5, 219-227.

Smith. F. Reversal of meaning as a variable in the trans-
formation of grammatical sentences. J. verb. Learn.
verb. Behav.. 1965, 4(1), 39-43.

Wiener, M. and Cromer. W. Reading and Reading Difficulty:
A Conceptual Analysis. Harvard Educational Review,

1967, 37. Number 4, pp. 620-643.

Winer. B.J. Statistical Principles in Experimental Design.
McGrawaHiII: 1962.

III
I.)
I"
L III
H
"
Ill|||l
"

5
7
1
2
9
6
1
3

1

3 mu)” mun

IIHIIUHIH