. 3) 9:3... 212:5. ,. ,;. 2. I. f: . ‘ .339“ x .A . , . . .3} 3!... Kr... . If... . . t. . .\ f. it. .25.... . 1.23.. .p 57...... . . . . I r... .7... . 5....t.r.5...ru.: 5 .2. . r3... 2.. .3. . ..~._m~v:.fi\ £52. ra-v.uh...u...«ir:r w _ a r. . =1. .. 3 7.5.1. E .. . w I: . .. .z .2. .. K. ‘ . . . . .2. .5. .3. 22...} _.. .2...» ............J....... .rz......=\ 1K“. 3.. . 5:1... v: T 1;. ..; ........:2...._Z.... .x....§a\.\ £3.22... . L2,... . 5.3.x . .. x. : @3393: ... .2. 5.9.5:: 2. Ur ... .1 .. 5.. . Ac : .2557. 1? k7»? .23.... . . ._:......v AG. 2.. «A. .4 2:3,. .. .. 5.... . . . .. . {3.51: .5 Z. 5‘: 51:12.»... A .2331: s 5.1.: ,. I... turn“... . _ z .15.. . sub. 5 .. 5 y .: v.7. . 5...} 19...}: 3...: v» \7 $28.: . . .5. .3: 2.1? r. . .2! . 2 21...}... r. .. a2. .. .. . . 5 .. .x . .2 :11: ..; 7721.... 3. >. fit... x :1... 141;...) 5:. y 52...... :3: .1. .17.. 2 2.. z. 571...: 2C .. 35......itx. . . 27.... fix 3 .2 J ‘5. 2.. 2 . 2 .7. 1...: . :15: .2512... . « x..=....11.=v. «if... .32.. 5;. v. :3; .1; 3.. I .1 .2}. ..L. 13.1.7711. r...» u ....:.,:£. :1“. 3 q. . .: P {7... . .1 s. r... : 7...”... . 5.1.11} 3. 2...: .3... LE. . 3e £2»... E. «.52. $1.... .1... ".4“ .1 Entry: I... 5 1 . . 2...... :742. x... .52: 2. .1. r... ,1 .5 ~ ‘1. 1.. .1. ..... .. 13......)2: :15). 3.... a. . A .:.. :12». . s. 3:... . .1511. . .1 3. 7... 7. .24.... :11»... ...,........1;r 1.3. 1...... .: L . 2.9.111...» .9. 316.... 5:» .. . . <. . .. 25.. . .u .z 7.3%. «349. .4 2 3., e. «.220: . 2.... . . Mun-r . x a .. .12: L3,... .3... A: :4 : . n. N... .. . .1 7.7.... .43... :2. .. ..4.;..bt.. .12... .. .. . .... .51. 3. . u. .3 I E. 3.. rrts...... é . 2...: , .1. m.....5..r 0 yrr . 12:!- u cmaraflw dish). 4 n2, .3, $5.3. . .:_;.!.....Lnrvcr.flmr. ' u a n. 1.... .(- 4.1: :13 4 d0 5.}. 51.31.4315... r o. 1 . 3.9.4.: .; . .3 . i.) 1. J .55.. 1,5? : .55.... 5...... at: z a. 5 . I .1. .115... 7:1. 5.. I... 7:)... .. : .fli .1: 3.... .. 7.2.2.5.. . I. f... 22.1.4153. . . ..v.,...:... .5... I. 7. . .z . 9..» :2; r1)...» 3... .3 5.5.5.5. Z 5.... i. i .. 5:..ng v (.3 x t: I. 5... wit...) . .11.. .. ....:. .... . .3... L .. a . .7. .: Ergtsr‘... . . . 5.. . ., f... 2.3... u » . x; 31.1.... . . . .4... 3. . £2... 55.3.4 . , u 1. I?) rthHJ . x ,r ., .. I. .K b.5525... 5.255. & gnu... m if 3:. 2...... i 2...... .1. r. r... ..v... r...;v .. 2. .. arr... $19....» .7... . . a... ,L... .r 7.»... . r i»... < I .: 2... 2.. a v 3.. Lib... .t....:.:_.m!.. .. . .4 , H,..xa...._.. .. .. . $.55. 4 .5... .r. . $0.426. Jr; 5.? n.4,.» 5...; pm... .1. .5... 5:2: .t.a:.L.o.. (fix/r .r Irz. w . W I: 2.9 .. .. . 7:7. 2.91. 5.... . :5. 1 J t. . :3. fix . 5...... .. .1... 5.2.3,. . v: . SN. 75 712.... 11.12.7151,. . .I.,......\...r.v.r..v.I; .1: :.....r 9 f5 Elf-.2431. . .121 a... . ( 5.}! . (1 .5). ,..(.;,. 1.2.1.31; .,./;.;r r f... , I 2va. :5... (If. . . 07639 LIBRA : Michigan Sta Le University This is to certify that the thesis entitled Cardiac and Somatic Concomitants of Response Activation and Response Inhibition presented by Terry Walter Allen has been accepted towards fulfillment of the requirements for 42191491— degree in #syeheleg} ABSTRACT CARDIAC AND SOMATIC CONCOMITANTS OF RESPONSE ACTIVATION AND RESPONSE INHIBITION By Terry Walter Allen Recent research has found that patterns of cardiac and somatic change are related to performance in a reaction time task. However, studies which have examined the generalization of these relationships across different tasks have had limited success. The relationship between patterns of physiological change and reaction time was studied using an Activation Task and an Activation-Inhibition Task with male college students. In contrast to previous investigations, the type of preparatory interval (fixed or variable) was covaried with task. Magnitude of heart rate deceleration was found to be related to reaction time for §s in both fixed preparatory interval Activation and Activation-Inhibition Task groups. Preperiod heart rate variance and Tonic heart rate variance reduction were related to reaction time for §s in both variable preparatory interval Activation and Activation-Inhibition Task groups. Tonic muscle activity was related to reaction time for §s in both fixed and variable preparatory interval Activation Task groups. In the fixed preparatory Activation-Inhibition Task, low Tonic muscle activity was related to successful response inhibition. Magnitude of Tonic heart rate variance reduction was related to successful response inhibition Terry Walter Allen for §s in the variable preparatory interval Activation- Inhibition Task group. The results of this study were discussed in terms of their implications for a physiological basis of attention. CARDIAC AND SOMATIC CONCOMITANTS OF RESPONSE ACTIVATION AND RESPONSE INHIBITION By Terry Walter Allen A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Psychology 1973 6m at (3.: ACKNOWLEDGMENTS This study was submitted by the author in partial fulfillment of the requirements for the degree of Doctor of PhiIOSOphy. The author would like to thank the members of the thesis committee: Drs. William Crano, Ellen Strommen, and Gordon WOOd. The author would like to especially thank Dr. Hiram.Fitzgerald, chairman of the thesis committee and adviser, for his advise and aid given during the course of the research and the guidance provided as part of the author's graduate training. Special thanks are due to Mitchell Leibowitz for his assistance in data collection. 11 TABLE OF CONTENTS List of tables....................................1v List of figures...................................v1 Introduction.......................................1 Method.............................................5 ResultS...........................................11 Discussion........................................u1 List of references................................48 Footnotes.........................................50 AppendixOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO0.0.051 iii Table LIST OF TABLES A representation of the eXperimental design prior to analysis for color preference. A representation of the experimental design following the analysis for color preference. Summary of the analysis of variance of reaction time as a function of preparatory interval (PI) and trials for §s in the Activation groups. Summary of the analysis of variance of reaction time as a function of preparatory interval (PI) and trials for SS in the Activation-Inhibition groups. The mean reaction times in milli- seconds for each Task group and Preparatory Interval (PI) group with their subsequent tgvalues. Error frequencies across five trial blocks (trial block = trials 1 - 5) for each eXperimental group. Summary of the analysis of variance of error frequency as a function of preparatory interval (PI) and trial blocks for SS in the Activation- Inhibition groups. Summary of the analysis of variance of mean heart rate as a function of task (T), preparatory interval (PI), Trials (Tr), and period (Per) for Ss in all eXperimental groups (Groups A—FI, A-VI, AI-FI, AI-VI). iv Page 12 15 16 17 19 20 21 Table Page 9. Summary of the analysis of variance 26 of heart rate as a function of trials (Tr), period (Per), and seconds (Sec) for gs in Group A-FI. 10. Summary of the analysis of variance 28 of heart rate as a function of trials (Tr), period (Per), and seconds (Sec) for gs in Group A—VI. 11. Summary of the analysis of variance of 29 heart rate as a function of trials (Tr), period (Per), and seconds (Sec) for SS in Group AI-FI. 12. Summary of the analysis of variance 32 of heart rate as a function of trials (Tr), period (Per), and seconds (Sec) for §s in Group AI-VI. 13. Summary of the analysis of variance 33 of mean heart rate variance as a function of task (T), preparatory interval (PI), trials (Tr), and period (Per) for §s in all eXper- imental groups (Groups A-FI, A-VI, AI-FI, and AI-VI). 1U. Summary of correlational analysis in 39 which antecedent physiological activity was correlated with reaction time performance for each eXperimental group. Figure LIST OF FIGURES A representation of the reaction time paradigm. Mean reaction times on each trial for Se in Activation groups A-FI and A-VI. Mean reaction times on each Activation trial for §S in Activation-Inhibition groups AI-FI and AI-VI. Mean heart rate on each trial for §s in Activation and Activation-Inhibi- tion tasks. Mean heart rate during each period for Ss in Activation and Activation- Inhibition tasks . Mean heart rate during each period for gs in all eXperimental groups for trials 1, 10, and 20. Mean heart rate during each period for SS in Group AI-FI. Mean heart rate variance during each period for §s in all eXperimental groups. Mean heart rate variance during each period for Ss in Activation and Activation4Inhibition tasks. vi Page 10 13 1L» 22 2M 25 31 34 36 INTRODUCTION Recent research has found that patterns of physio- logical change used to index an individual's attentional behavior are related to task performance. This relation- ship between physiological responding and performance has been most frequently examined using a reaction time task. In a reaction time task, a signal (warning signal) is presented, the purpose of which is to prepare S for a second stimulus (response signal), to which § must respond as quickly as possible. The time between the onset of the warning signal and the response signal is called the preparatory interval. Using a reaction time task, investigators have been chiefly interested in studying relationships between cardiac and somatic activity and reaction time performance. Several investigators have found that a large heart rate deceleration Just prior to the response signal was associated with a fast reaction time (Krupski, 1971; Lacey, 1967; Schwartz & Higgins, 1971). This relationship appears to be specific to those reaction time tasks employing a fixed preparatory interval in contrast to those employing variable preparatory intervals. In addition, Porges (1972) recently has examined heart rate variability as a component of attention and its relationships to reaction time performance. 1 2 In this study, it was found that mean heart rate variance reduction in anticipation to the response signal and mean pretrial variance were related to reaction time. However, this result was true only for §s in a variable preparatory interval condition and not for §s in a fixed preparatory interval condition. In addition to these heart rate indexes of attention and subsequent reaction time performance, anticipatory muscle activity has been found to be related to reaction time performance. Using electromyographic (ENG) recording of chin muscle activity, Jennings, Averill, Opton, and Lazarus (1971) found that increases in muscle activity prior to responding were associated with faster reaction times. Grossman, Fitzgerald, and Porges (1971) also found that increases in forearm circumference for males (task relevant muscles) were correlated with fast reaction times. On the other hand, increases in forearm circumference were correlated with slow reaction times for females. The results of these studies suggest that both heart rate deceleration and variability as well as muscle activity are useful indexes of attentional involvement and sub- sequent performance. However, investigators which have studied the generality of relationships between physiological and performance in variations of the reaction time task or other more complex tasks have reported little success. Jennings 3542;. (1971) examined the relationships between heart rate deceleration, muscle activity, and reaction time 3 under various task conditions in which §s were required to perform more than one response or underwent stress. Although some relationship between muscle activity and reaction time was observed across tasks, relationships between heart rate deceleration and reaction time were limited. Moreover, Stroufe (1971) compared children's performance on a simple fixed preparatory interval reaction time task and on a task which required them to respond or not respond to a per- ceptual signal. In both tasks, Stroufe found that magnitude and temporal preciseness of heart rate deceleration were related to reaction time. However, Stroufe did not find a significant relationship between inhibition performance (errors) and heart rate deceleration. The above findings have important implications for hypotheses which prOpose a physiological basis for attention. If heart rate deceleration or other physiological responses are to be regarded as a strong index of the organism's attention, then these responses should be related to per- formance in other tasks. In particular, the results of these studies suggest that heart rate deceleration may be limited in its value as an index of the individual attention when used in tasks other than the simple reaction time task. However, the failure to obtain this relationship between patterns of physiological change and performance across tasks may be due to a failure to consider the type of preparatory interval employed in a reaction time task or variations of this task. In all of the previous studies L; reviewed, preparatory interval in combination with task demands has not been systematically varied. This is an important omission since previous research has shown that heart rate and somatic responses are affected considerably by the nature (fixed or variable) of the preparatory interval employed in the task (Krupski, 1971; Lacey, 1967; Porges, 1972). Consequently, a more conclusive test of generality of these responses as predictors of performance would be an eXperiment in which the effects of type of preparatory interval in conjunction with task demands on heart rate and somatic activity were examined. The present eXperiment was designed to study the relationship between physiological response patterns (cardiac and somatic components of attention) and reaction times under various task conditions. Preparatory interval (fixed or variable) was covaried with task. Two tasks were used in the present study: a simple reaction time task (Activation Task) and an Activation-Inhibition Task patterned after the one employed by Luria (1961) in which g is required to respond to one signal and inhibit the same response to another signal. It was hypothesized that those cardiac and somatic response patterns related to reaction time performance in a simple reaction time task would also be related to successful response inhibition and reaction time for §s in the Activation-Inhibition Task. Method SubJects Subjects were #8 college-age-males acquired from the introductory psychology course offered at Michigan State University and from the East Lansing area. Subjects recruited from Michigan State University received extra course credit for their participation in the eXperiment while those obtained from the East Lansing community were paid. Six §s were discarded because of mechanical failure of equipment. Apparatus Stimuli. The stimulus presentation apparatus consisted of a gray panel (12" x 12" x 6") on which amber, blue, and white 24v. lights were mounted. These lights were presented at a distance of h ft. from g at eye level. During the Activation task, onset of the amber light warned S to pre- pare to respond while its offset signaled §,to respond. In the Activation plus Inhibition condition, the amber light onset signaled § to respond or inhibit his response. The onset of the blue or white lights was simultaneous with the offset cf the amber light and signaled§ to respond or not to respond. The presentation of these stimuli (stimulus duration and intertrial interval) was controlled by Hunter timers. The eXperiment was conducted in a sound-attentated room with a temperature of 70 degrees F. The ambient noise level was 51 db. Response measurement. Reaction time (RT) was measured using a bulb connected to a vacuum switch (Grossman‘g£.al., 1970). Reaction time was registered in milliseconds on a Standard electric clock. PhysiologiCal responses were recorded on a four-channel Grass polygraph, Model P7. Beckman biOpotential silver-silver chloride electrodes were used to record EKG and electromyographic activity while Beckman biopotential paste was used as the electrolyte. Heart rate (HR), unintegrated electromyographic responses (EMG), and forearm circumference (FC) were recorded. Heart rate was recorded with a Grass Model 7PhA tachograph. A Grass Wide-Band Preamplifier and Integrator, Model 7P3B, was used to measure EMG amplitude. Forearm circumference was recorded using a Parks Electronic four-inch mercury strain gauge (Grossman et al., 1970). Changes in circum- ference were measured by a Grass Low-Level DC Pre-Amplifier, Model 7P1A. Procedure Subjects were randomly assigned to Groups 1 through 6 when they arrived at the experimental room (See Table l). The subject was seated and EKG and EMG recording sites were prepared. EKG electrodes were attached to the right leg near the ankle (g), to the left arm (+), and to the right arm (-). EMG electrodes were placed on the right forearm one-third the distance between the medial epicondyle of the humerus end the styloid process of the radius; and the second electrode was placed two inches in the distal A representation of the experimental for color preference. TABLE 1 design prior to analysis Group Task Preparatory Stimulus Color of Interval Respond Signal 1 Activation Fixed White or Blue 2 Activation Variable White or Blue 3 Activation- Fixed White Inhibition Blue u Activation— Fixed Blue Inhibition (counterbalance) White 5 Activation- Variable White Inhibition Blue 6 Activation- Variable Blue Inhibition (counterbalance) White 8 direction from the first electrode along the same line (Venables & Martin, 196h). The ground electrode was located on the bicep of the same arm. The strain gauge was placed between the two EMG electrode sites on the forearm. Following attachment of recording electrodes and the strain gauge, the task instructions were read to g. Subjects in Groups 1 and 2 were told to watCh for the appearance of the amber light. When the amber light disappeared, §s were told to respond as rapidly as possible when it disappeared and the blue light came on. Subjects were also told to not respond when the amber light dis- appeared and the white light came on. Subjects in Groups 4 and 6 were given the same instructions except that the Activation and Inhibition stimulus values used for Groups 3 and 5 were reversed. The polygraph was then calibrated. Following calibration, lights in the experimental room were dimmed and S was informed via an intercom that the experiment was about to begin. At the completion of the experiment, §s were informed via the intercom that the eXperiment had ended. Each § received twenty trials during the eXperimental session. For Groups 1 and 2, all twenty trials were Activation trials. During the experimental period for Groups 3 through 6, ten trials were Activation trials and ten trials were Inhibition trials. A predetermined random schedule ordered the appearance of the Activation and Inhibition trials. All groups had the same TTI which varied among #5, 60, and 75 seconds. The length of the 9 fixed ISI condition (Groups 1, 3, and 4) was 32 seconds. For the variable ISI condition, the ISI length varied among 16, 22, and 28 seconds (Groups 2, 5, and 6) with a mean of 22 seconds. Quantification of Data Reaction time, HR, EMG, and FC were scored for each of the twenty trials during the eXperimental period. Each trial was divided into four periods for analysis (see Figure 1): (a) a Preperiod consisting of 8 seconds prior to the onset of the amber light (warning signal); (b) a Phasic period of 8 seconds immediately following the onset of the amber light; (0) a 22319 period consisting of 8 seconds prior to the onset of the blue or white lights; and (d) a Response period of 8 seconds immediately following the onset of the blue or white lights (see Porges, 1972). Reaction.£im§. Subjects' RT was defined as the latency between offset of warning signal and the bulb squeeze. Response errors. Those inhibition trials in which S responds or those Activation trials in which §_failed to respond-~were recorded. £3233 5233. Subject's HR (beats per minute, bpm) was scored directly from the printed output from the polygraph. A HR score was obtained for each of the 8 seconds contained within the four periods of analysis. If more than one heart beat occurred during one second, only the last beat to occur was scored. 10 .awvmumm Tip. concurs of mo cowumuccmmuaca < A sham: on» m 08 a “ co» m can m .935 .933 became: 05:3; voted 3:698... eaten. Bach voted 032.5 3:395 .0225 329.305. 11 Egg. EMG amplitude was computed by counting the number of spike potentials occurring during each of the four periods of analysis. A spike potential was defined as a change in polarity greater than or equal to 50 microvolts. F91. Changes in §'s FC were obtained by measuring millimeters of pen deflection from base line for each second during the four periods of analysis. Results Analysis of the data was divided into three phases: (a) a reaction time phase; (b) a physiological activity phase; and (c) a relationship between physiological activity and RT phase. Analysis of reaction time con- sisted of examining the eXperimental effects on reaction time and error frequency. Experimental effects on heart rate and muscle activity were examined during the analysis of physiological activity. Finally, the relationship between physiological state and performance was examined for each eXperimental condition. Before considering the results of these analyses, one preliminary analysis was done. Although there was no a priori reason to suspect that the color of the response signal would influence performance, the data were analyzed for possible color preferences. Reaction times to the blue light were compared with those to the white light yielding no significant differences as a function of response signal color (t = 1.03, df = 718, p )'O.20). Consequently for the 12 rest of the data analysis, Groups 3 and 4 were combined to form Group AI-FI and Groups 5 and 6 were combined to form Group AI-VI (see Table 2). TABLE 2 A representation of the eXperimental design following the analysis for color preference. Group Task Preparatory Interval l (A-FI) Activation Fixed Interval 2 (A-VI) Ictivation Xariable Interval 3 + u (AI-F1) Activation-Inhibition Fixed Interval 5 + 6 (AI~VI) ActivationfiInhibition yariable Interval Reaction Time Reaction time data for all conditions were plotted. The plotting of RT data for the Activation conditions appear in Figure 2 while those for the Activation-Inhibition con- ditions appear in Figure 3. An analysis of variance was performed on the RT data for the Activation groups while a separate analysis of variance was performed on RT data for the Activation-Inhibition groups. Results of this analysis (see Table 3) for Group A-FI and Group A-VI indicate that RT performance improved across trials. Results of the analysis of variance for Groups AI-FI and AI-VI (see Table 4) also indicated that RT performance improved across trials. 13 .H,I< vac ~m|< wmzouw cofium>fiuu< aw mm sou Hews» zone :: 195w» cowuomwa cam: L .V shaman 9.55. P ”P. b w. h c?— » NP— » Ch. > m p m > W W I o Aroom town finxue. x). x», a \ \‘ll‘l \ 0K \\ < / x 4/1 \ , x) yonv , \ (IA r , a , x c 1000 100m 1com I 5.4 doomo 9‘ r74 anomo room 3N”. NOLLOVBB NVBW SONOOBSIT'IIW NI MEAN REACTION TIME IN MILLISECONDS 650* 600* 550* 500- 450‘ 400 350‘ 300: I ‘V 1U GROUP AI-VI 0-w- GROUP AI-VI H O 'T Figure 3. E T j V fl V T Sissrasuo TRIALS Mean reaction time on each Activation trial for gs in Activation- lnhibition groups AI-FT and Al-VI. 15 TABLE 3 Summary of the analysis of variance of reaction time as a function of preparatory interval (PI) and trials for §s in the Activation groups. Source SS df MS F p PI 5333.33 1 5333.33 0.08 (0.20 Error 1568604.66 22 71300.21 Trials 185824.76 19 9727.62 1.92 (0.05 Trials X PI 10751h.15 19 5658.6“ 1.12 ‘<1.20 Error 211532h.69 #18 5060.59 Total 398160l.59 16 TABLE 4 Summary of the analysis of variance of reaction time as a function of preparatory interval (PI) and trials for §s in the Activation-Inhibition groups. Source SS df MS F p PI 88051.70 1 88051.70 1.54 '