Perceptual and Motor Skills, 1975,41, 395-399. @ Perceptual and Motor Skills 1975
EFFECTS OF TASK DIFFICULTY ON SUBSEQUENT PREFERENCE FOR VISUAL COMPLEXITY1 HAL R. ARKES AND PATRICIA CLARK Ohio University
.
Summary.-Ss rated o n a 1 to 1 3 Likert scale their preference for four environmental scenes which differed in complexity. Ss were then given 30 sec. to solve each of 10 3-, 5-, or 7-letter single-solution anagrams. Finally the Ss rated four new slides which were of the same complexity level as the slides seen during the first session. Change scores were calculated by subtracting the preference ratings at each complexity level during the first session from the rating at the corresponding complexity level during the second session. The Anagram Complexity x Slide Complexity interaction was significant; the complexity of the slides showing the most positive change scores was inversely related to the complexity of the anagrams attempted. The results were discussed in terms of optimal level of srimulation theories.
During the last fifteen years a substantial amount of research has been done in the area of exploratory behavior and stimulus preference. Early theorists (Dember & Earl, 1957; Berlyne, 1960; Walker, 1964) all posited some o p timal level of arousal or complexity. The organism would always strive to adjust his present level of stimulation in order to attain his optimal level. Recent investigations of these optimal level theories have focused on a previously neglected area of research-the effect of prechoice stimulation on subsequent stimulus preference. For example, using colored slides of tourist scenes, Berlyne and Crozier (1971) found that Ss' proportion of choices of more complex visual stimulation varied inversely with the complexity of the prechoice visual stimulus. Similarly, Leckart, Levine, Goscinski, and Brayman ( 1970) and Leckart, Glanville, Hootstein, Keleman, and Yaremko ( 1972 ) found that looking time ar random shapes and other stimuli varied directly with the amount of prechoice visual deprivation. These findings can all be explained by optimal level of stimulation theories. Complex prechoice stimulation might represent stimulation above S's optimal level. S would then be expected to decrease his choice of complex stimuli and increase his choice of simpler ones, thus striving to return to the optimal level. Similarly, the Leckart, et al. (1972) study showed that stimulus deprivation, which represents a suboptimal level of stimulation, resulted in increased looking time at subsequent stimulation as S tried to attain an optimal level of stimulation. A puzzling feature of the Leckart, et al. (1972) study was that stimulus deprivation increased looking time toward 'We wish to thank Chonita Spanja lor her able assistance. A portion of this paper was presented at the Midwestern Psychological Convention, Chicago, 1974. Reprint requests should be sent to Hal R. Arkes, Ohio University, Department of Psychology, Athens, Ohio 45701.
396
H. R. ARKES ~r P. CLARK
all levels of stimulus complexity. It might be expected that S would look longer at complex rather than simple stimuli in order to correct the suboptimal level of stimulation induced by the deprivation period. However, as stimulus deprivation increased, looking times at all levels of stimulus complexity increased approximately equally. One of the purposes of the present study was to investigate further whether various levels of prechoice stimulation do have a differential effect on subsequent preference for stimuli of different complexity levels. A second purpose of this study was to determine whether prechoice stimulacion other than the types of visual stimuli used in prior studies can influence subsequent complexity preference. Additional substantiation of optimal level theories will be more meaningful if several different types of stimulation are used in relevant experiments. In this study prechoice stimulation was varied by giving anagrams of different difficulty to three groups. It was hypothesized that Ss who worked on difficult anagrams would show decreased preference for complex stimuli and increased preference for simple stimuli. Ss who worked on easy anagrams would show increased preference for complex stimuli and decreased preference for simple ones.
Forty-eight female undergraduates served as Ss and received course credit for their participation. They were divided into three groups of 16 each. Procedure
The visual stimuli were black and white slides of pictures from photography magazines and ecology texts. The slides represented environmental scenes (mountains, snowdrifts) and still life photos (a pitcher on a table, a vase). The eight stimuli used in the experiment were chosen from a larger group of 39 slides, each of which was shown to two university classes for 15 sec. One class rated the slides for "liking" on a 1 to 1 3 Likert scale, "1" being "really dislike" and "13" being "really like." The other class rated the slides for "complexity" on a scale of 1 to 13, with "1" being "extremely simple" and "13" being "extremely complex." The eight slides chosen from this set were divided into two sets of four each, each set containing slides varying from high to low complexity. The two sets were as similar as possible with regard to their complexity and preference ratings. All Ss were tested in pairs in a dark room containing four Carousel projectors and four screens arranged in a semi-circle 2.5 m in front of the two Ss. The female E sat between Ss and operated the projectors with remote control switches. The experiment began by simultaneously projecting one slide on each of the four screens for 1 min. Half of the Ss saw set A; half saw set B. The left-
PREFERENCE FOR COMPLEXITY
397
to-right order of the slides within the semi-circle of screens was completely counterbalanced. During chis 1-min. viewing session, Ss rated each of the slides on a 1 to 13 preference scale with 1 being the most negative score. Immediately following this viewing session, Ss were given 10 3-, 5-, or 7-letter single-solution anagrams to solve. All anagrams were scrambled versions of words occurring at least 35 times per million according to the Thorndike-Lorge (1944) norms. Each anagram was presented for 30 sec. on the second movie screen from the left. Ss recorded their answers on sheets provided. Immediately following presentation of the last anagram, Ss were shown either set A or B, whichever was not seen during the first viewing session. Once again Ss saw all four slides presented simultaneously for 1 min. Each slide was rated on a 1 to 13 preference scale. Once again the left-to-right order within the semi-circle was counterbalanced. It should be pointed out that a possible control group viewing no anagrams was deliberately omitted from this study. If such a group spent the 5 min. between slide sets doing no assigned task, the complexity of their behavior would be uncontrolled. Very substantial pilot work has indicated that sitting alone for 5 min. in a darkened room is rated by the S as not as boring as viewing simple stimuli. Therefore, doing no anagrams would not constitute a control of "nocomplexity." Giving the two slide-viewing sessions to S with no time between sessions would also constitute an inadequate control. Such a group would differ from the experimental groups in two ways, time duration and complexity of activity. The effect of complexity therefore could not be unequivocally assessed. Since the present experiment hypothesizes different effects of anagram complexity at various visual complexity levels, a control group assessing whether the anagram task has any effect is unnecessary.
RESULTS First, it is necessary to determine if the experimental manipulation, the anagrams, truly represented different levels of cognitive complexity. The mean numbers of anagrams solved by the 3-, 5-, and 7-letter groups were 9.94, 4.19, and 2.3 1 respectively. Since all but one S in the 3-letter anagram group solved all 10 anagrams, there was almost no variance in this group; therefore an analysis of variance including this group would be inappropriate. Since neither the 5- nor the 7letter groups' distribution of number of anagrams correctly solved overlapped the 3-letter group's distribution, the 3-letter group therefore differed significantly from the other two groups. The number of anagrams solved by the 5and 7-letter groups differed significantly ( t = 1.84, df = 30, 9 < .05). Thus the three groups of anagrams truly did differ in difficulty. Next, the effect of the anagram session on subsequent viewing of four cornplexity levels was assessed. Four change scores were calculated for each S by
H. R. ARKES 8c P. CLARK
398
subtracting the rating at each complexity level during the first session from the rating at the corresponding complexity level during the second session. A threeway analysis of variance was done with complexity of slide as a within-subjects factor and complexity of anagram and order (set A or B seen first) as betweensubjects factors. As predicted, the complexity of slide X complexity of anagram interaction was significant ( F = 2.52, df = 6/126, p < .025). Those who attempted difficult anagrams increased their preference for the least complex slide, while those who attempted easy anagrams decreased their preference for the least complex slide. Those who attempted easy anagrams increased their preference for the most complex slide more than did those who attempted difficult anagrams. The 5-letter anagram group surprisingly showed decreased preference at both the highest and lowest visual complexity levels.
'I W L7
3 -
u
-
W
I
2
5: 0 Z
'I -
,
:: P
E"
0 -
W
g
---
3- LETTERS 5-LETTERS 7-LETTERS --,---
-I -2
C
LEAST
MOST
SLIDE COMPLEXITY
Frc. 1. Change in preference ratings as a function of slide complexity and anagram length
Also significant, but not predicted, was the interaction of order X slide complexity ( F = 6.99, df = 3/126, p < .01). When set A was seen first, the change scores for the most complex slide were lower and for the least complex slide were higher than when set B was seen first. This undoubtedly reflects a difference in liking, regardless of anagram condition, between the polar members of the two sets. The reported difference in Fig. 1 between the 3- and 7-letter groups concerning their preference for the most and least complex slides respectively was apparent in both the A-B and B-A groups. Thus, this effect strongly superseded any idios~ncraciesof one or more of the slides in each set. This study was designed to answer two related questions: could visual complexity preference be influenced by the complexity of a different type of prior
PREFERENCE FOR COMPLEXITY
399
cognitive experience, and secondly, is the visual complexity spectrum differentially affected by the complexity of the prior task? The interaction of complexity of slide X complexity of anagram answered both questions affirmatively. As optimal level of stimulation theories would predict, the complexity level of the slides showing the greatest positive change scores was inversely related to the complexity level of the prior anagram task. The $letter group showed the greatest positive change for the most complex slides, the 5-letter group for the third most complex slide, and the 7-letter group for the two simplest slides. Those above their optimal complexity level preferred simple slides, while those below their optimum preferred complex ones. Despite the regularity of the above findings, the preference exhibited by the 5-letter groups for the rwo extreme slides was puzzling. It is not readily apparent why the preference of the 5-letter group for these two slides is not between the preference exhibited by the 3- and 7-letter groups. Finally, it is important to note certain implications of the finding that a cognitive task can influence visual complexity preference. Environmental psychologists have been investigating the relationship between one's cognitive state and one's perception of the environment (Lynch, 1965; Wohlwill, 1966). Whereas many theorists have speculated that environmental complexity could influence one's cognitive state, the present study turns the tables: one's cognitive state can influence subsequent preference for environmental complexity. REFERENCES
BERLYNE, D. E. Conjlict, arousal, and curiority. New York: McGraw-Hill. 1960. BERLYNE,D. E., & (~ROZIER, J. B. Effects of complexity and prechoice stimulation on exploratory choice. Perception O P~~chophysics,1971, 10, 242-246. DEMBER. W. N., & EARL, R. W. Analysis of exploratory, manipulatory, and curiosity behaviors. Psychological Review, 1957, 50, 514-518. LECKART, B. T., GLANVILLB, B., HOOTSTEIN, E., KELEMAN,K., & YAREMKO, R. M. Looking time, stimulus complexity, and the perceptual deprivation effect. Psychonomic Science, 1972, 26, 107-108. LECKART, B. T., LEVINE,J. R., GOSCINSKI, C., & BRAYMAN, W. Duration of attention: the perceptual deprivation effect. Perception & Psychophysics, 1970, 7, 163-164. LYNCH,K. The city as environment. In Cities: a Scientific Arnerican book. New York: Knopf. 1965. Pp. 192-201. THORNDIKE, E. L., & LORGE, I. The teacher's word book o f 30,000 words. New York: Columbia Univer., 1944. WALKER, E. L. Psychological complexity as a basis for a theory of motivation and choice. In D. Levine (Ed.), Nebraska symposium on motivation. Lincoln: Uoiver. of Nebraska Press, 1964. Pp. 47-95. WOHLWILL,J. F. The physical environment: a problem for a psychology of stimulation. Journal o f Social Irruer, 1966, 22(4), 29-38.
Accepted June 23, 1975.
EFFECTS OF TASK DIFFICULTY ON SUBSEQUENT PREFERENCE FOR VISUAL COMPLEXITY1 HAL R. ARKES AND PATRICIA CLARK Ohio University
.
Summary.-Ss rated o n a 1 to 1 3 Likert scale their preference for four environmental scenes which differed in complexity. Ss were then given 30 sec. to solve each of 10 3-, 5-, or 7-letter single-solution anagrams. Finally the Ss rated four new slides which were of the same complexity level as the slides seen during the first session. Change scores were calculated by subtracting the preference ratings at each complexity level during the first session from the rating at the corresponding complexity level during the second session. The Anagram Complexity x Slide Complexity interaction was significant; the complexity of the slides showing the most positive change scores was inversely related to the complexity of the anagrams attempted. The results were discussed in terms of optimal level of srimulation theories.
During the last fifteen years a substantial amount of research has been done in the area of exploratory behavior and stimulus preference. Early theorists (Dember & Earl, 1957; Berlyne, 1960; Walker, 1964) all posited some o p timal level of arousal or complexity. The organism would always strive to adjust his present level of stimulation in order to attain his optimal level. Recent investigations of these optimal level theories have focused on a previously neglected area of research-the effect of prechoice stimulation on subsequent stimulus preference. For example, using colored slides of tourist scenes, Berlyne and Crozier (1971) found that Ss' proportion of choices of more complex visual stimulation varied inversely with the complexity of the prechoice visual stimulus. Similarly, Leckart, Levine, Goscinski, and Brayman ( 1970) and Leckart, Glanville, Hootstein, Keleman, and Yaremko ( 1972 ) found that looking time ar random shapes and other stimuli varied directly with the amount of prechoice visual deprivation. These findings can all be explained by optimal level of stimulation theories. Complex prechoice stimulation might represent stimulation above S's optimal level. S would then be expected to decrease his choice of complex stimuli and increase his choice of simpler ones, thus striving to return to the optimal level. Similarly, the Leckart, et al. (1972) study showed that stimulus deprivation, which represents a suboptimal level of stimulation, resulted in increased looking time at subsequent stimulation as S tried to attain an optimal level of stimulation. A puzzling feature of the Leckart, et al. (1972) study was that stimulus deprivation increased looking time toward 'We wish to thank Chonita Spanja lor her able assistance. A portion of this paper was presented at the Midwestern Psychological Convention, Chicago, 1974. Reprint requests should be sent to Hal R. Arkes, Ohio University, Department of Psychology, Athens, Ohio 45701.
396
H. R. ARKES ~r P. CLARK
all levels of stimulus complexity. It might be expected that S would look longer at complex rather than simple stimuli in order to correct the suboptimal level of stimulation induced by the deprivation period. However, as stimulus deprivation increased, looking times at all levels of stimulus complexity increased approximately equally. One of the purposes of the present study was to investigate further whether various levels of prechoice stimulation do have a differential effect on subsequent preference for stimuli of different complexity levels. A second purpose of this study was to determine whether prechoice stimulacion other than the types of visual stimuli used in prior studies can influence subsequent complexity preference. Additional substantiation of optimal level theories will be more meaningful if several different types of stimulation are used in relevant experiments. In this study prechoice stimulation was varied by giving anagrams of different difficulty to three groups. It was hypothesized that Ss who worked on difficult anagrams would show decreased preference for complex stimuli and increased preference for simple stimuli. Ss who worked on easy anagrams would show increased preference for complex stimuli and decreased preference for simple ones.
Forty-eight female undergraduates served as Ss and received course credit for their participation. They were divided into three groups of 16 each. Procedure
The visual stimuli were black and white slides of pictures from photography magazines and ecology texts. The slides represented environmental scenes (mountains, snowdrifts) and still life photos (a pitcher on a table, a vase). The eight stimuli used in the experiment were chosen from a larger group of 39 slides, each of which was shown to two university classes for 15 sec. One class rated the slides for "liking" on a 1 to 1 3 Likert scale, "1" being "really dislike" and "13" being "really like." The other class rated the slides for "complexity" on a scale of 1 to 13, with "1" being "extremely simple" and "13" being "extremely complex." The eight slides chosen from this set were divided into two sets of four each, each set containing slides varying from high to low complexity. The two sets were as similar as possible with regard to their complexity and preference ratings. All Ss were tested in pairs in a dark room containing four Carousel projectors and four screens arranged in a semi-circle 2.5 m in front of the two Ss. The female E sat between Ss and operated the projectors with remote control switches. The experiment began by simultaneously projecting one slide on each of the four screens for 1 min. Half of the Ss saw set A; half saw set B. The left-
PREFERENCE FOR COMPLEXITY
397
to-right order of the slides within the semi-circle of screens was completely counterbalanced. During chis 1-min. viewing session, Ss rated each of the slides on a 1 to 13 preference scale with 1 being the most negative score. Immediately following this viewing session, Ss were given 10 3-, 5-, or 7-letter single-solution anagrams to solve. All anagrams were scrambled versions of words occurring at least 35 times per million according to the Thorndike-Lorge (1944) norms. Each anagram was presented for 30 sec. on the second movie screen from the left. Ss recorded their answers on sheets provided. Immediately following presentation of the last anagram, Ss were shown either set A or B, whichever was not seen during the first viewing session. Once again Ss saw all four slides presented simultaneously for 1 min. Each slide was rated on a 1 to 13 preference scale. Once again the left-to-right order within the semi-circle was counterbalanced. It should be pointed out that a possible control group viewing no anagrams was deliberately omitted from this study. If such a group spent the 5 min. between slide sets doing no assigned task, the complexity of their behavior would be uncontrolled. Very substantial pilot work has indicated that sitting alone for 5 min. in a darkened room is rated by the S as not as boring as viewing simple stimuli. Therefore, doing no anagrams would not constitute a control of "nocomplexity." Giving the two slide-viewing sessions to S with no time between sessions would also constitute an inadequate control. Such a group would differ from the experimental groups in two ways, time duration and complexity of activity. The effect of complexity therefore could not be unequivocally assessed. Since the present experiment hypothesizes different effects of anagram complexity at various visual complexity levels, a control group assessing whether the anagram task has any effect is unnecessary.
RESULTS First, it is necessary to determine if the experimental manipulation, the anagrams, truly represented different levels of cognitive complexity. The mean numbers of anagrams solved by the 3-, 5-, and 7-letter groups were 9.94, 4.19, and 2.3 1 respectively. Since all but one S in the 3-letter anagram group solved all 10 anagrams, there was almost no variance in this group; therefore an analysis of variance including this group would be inappropriate. Since neither the 5- nor the 7letter groups' distribution of number of anagrams correctly solved overlapped the 3-letter group's distribution, the 3-letter group therefore differed significantly from the other two groups. The number of anagrams solved by the 5and 7-letter groups differed significantly ( t = 1.84, df = 30, 9 < .05). Thus the three groups of anagrams truly did differ in difficulty. Next, the effect of the anagram session on subsequent viewing of four cornplexity levels was assessed. Four change scores were calculated for each S by
H. R. ARKES 8c P. CLARK
398
subtracting the rating at each complexity level during the first session from the rating at the corresponding complexity level during the second session. A threeway analysis of variance was done with complexity of slide as a within-subjects factor and complexity of anagram and order (set A or B seen first) as betweensubjects factors. As predicted, the complexity of slide X complexity of anagram interaction was significant ( F = 2.52, df = 6/126, p < .025). Those who attempted difficult anagrams increased their preference for the least complex slide, while those who attempted easy anagrams decreased their preference for the least complex slide. Those who attempted easy anagrams increased their preference for the most complex slide more than did those who attempted difficult anagrams. The 5-letter anagram group surprisingly showed decreased preference at both the highest and lowest visual complexity levels.
'I W L7
3 -
u
-
W
I
2
5: 0 Z
'I -
,
:: P
E"
0 -
W
g
---
3- LETTERS 5-LETTERS 7-LETTERS --,---
-I -2
C
LEAST
MOST
SLIDE COMPLEXITY
Frc. 1. Change in preference ratings as a function of slide complexity and anagram length
Also significant, but not predicted, was the interaction of order X slide complexity ( F = 6.99, df = 3/126, p < .01). When set A was seen first, the change scores for the most complex slide were lower and for the least complex slide were higher than when set B was seen first. This undoubtedly reflects a difference in liking, regardless of anagram condition, between the polar members of the two sets. The reported difference in Fig. 1 between the 3- and 7-letter groups concerning their preference for the most and least complex slides respectively was apparent in both the A-B and B-A groups. Thus, this effect strongly superseded any idios~ncraciesof one or more of the slides in each set. This study was designed to answer two related questions: could visual complexity preference be influenced by the complexity of a different type of prior
PREFERENCE FOR COMPLEXITY
399
cognitive experience, and secondly, is the visual complexity spectrum differentially affected by the complexity of the prior task? The interaction of complexity of slide X complexity of anagram answered both questions affirmatively. As optimal level of stimulation theories would predict, the complexity level of the slides showing the greatest positive change scores was inversely related to the complexity level of the prior anagram task. The $letter group showed the greatest positive change for the most complex slides, the 5-letter group for the third most complex slide, and the 7-letter group for the two simplest slides. Those above their optimal complexity level preferred simple slides, while those below their optimum preferred complex ones. Despite the regularity of the above findings, the preference exhibited by the 5-letter groups for the rwo extreme slides was puzzling. It is not readily apparent why the preference of the 5-letter group for these two slides is not between the preference exhibited by the 3- and 7-letter groups. Finally, it is important to note certain implications of the finding that a cognitive task can influence visual complexity preference. Environmental psychologists have been investigating the relationship between one's cognitive state and one's perception of the environment (Lynch, 1965; Wohlwill, 1966). Whereas many theorists have speculated that environmental complexity could influence one's cognitive state, the present study turns the tables: one's cognitive state can influence subsequent preference for environmental complexity. REFERENCES
BERLYNE, D. E. Conjlict, arousal, and curiority. New York: McGraw-Hill. 1960. BERLYNE,D. E., & (~ROZIER, J. B. Effects of complexity and prechoice stimulation on exploratory choice. Perception O P~~chophysics,1971, 10, 242-246. DEMBER. W. N., & EARL, R. W. Analysis of exploratory, manipulatory, and curiosity behaviors. Psychological Review, 1957, 50, 514-518. LECKART, B. T., GLANVILLB, B., HOOTSTEIN, E., KELEMAN,K., & YAREMKO, R. M. Looking time, stimulus complexity, and the perceptual deprivation effect. Psychonomic Science, 1972, 26, 107-108. LECKART, B. T., LEVINE,J. R., GOSCINSKI, C., & BRAYMAN, W. Duration of attention: the perceptual deprivation effect. Perception & Psychophysics, 1970, 7, 163-164. LYNCH,K. The city as environment. In Cities: a Scientific Arnerican book. New York: Knopf. 1965. Pp. 192-201. THORNDIKE, E. L., & LORGE, I. The teacher's word book o f 30,000 words. New York: Columbia Univer., 1944. WALKER, E. L. Psychological complexity as a basis for a theory of motivation and choice. In D. Levine (Ed.), Nebraska symposium on motivation. Lincoln: Uoiver. of Nebraska Press, 1964. Pp. 47-95. WOHLWILL,J. F. The physical environment: a problem for a psychology of stimulation. Journal o f Social Irruer, 1966, 22(4), 29-38.
Accepted June 23, 1975.
