Experimental Aging Research
ISSN: 0361-073X (Print) 1096-4657 (Online) Journal homepage: http://www.tandfonline.com/loi/uear20
Differences in the sequential integration of form as a function of age and interstimulus interval Donald W. Kline & Gary Baffa To cite this article: Donald W. Kline & Gary Baffa (1976) Differences in the sequential integration of form as a function of age and interstimulus interval, Experimental Aging Research, 2:4, 333-342, DOI: 10.1080/03610737608257990 To link to this article: http://dx.doi.org/10.1080/03610737608257990
Published online: 27 Sep 2007.
Submit your article to this journal
Article views: 6
View related articles
Citing articles: 2 View citing articles
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=uear20 Download by: [Deakin University Library]
Date: 07 November 2015, At: 21:17
Downloaded by [Deakin University Library] at 21:17 07 November 2015
DIFFERENCES IN THE SEQUENTIAL INTEGRATION OF FORM AS A FUNCTION OF AGE AND INTERSTIMULUS INTERVAL
DONALD W. KLINE
GARY BAFFA
Psychology Defmtment University of Notre Dame Notre Dame, Jndiana 46556
Andrus Gerontology Center University of Southern California Los Angeles, California 90007
Kline, D. W., & Baffa, G. Differences in the sequential integration of form. Experimental Aging Research, 1976, 2 (4), 333-343. The sequential integration of visual stimuli was studied in 12 young (18-28 years) and 12 old (51-62 years) men and women. Stimuli were constructed as corresponding word halves and inter-stimulos intervals (ISI’s) of varied duration (0-150 msec.) were introduced between the presentations of the two halves. Correct word identification was found to be an inverse function of IS1 for both young and old subjects. T h e number of stimuli correctly identified was significantly lower for the older subjects a t all levels of ISI. None of the interactions of age, sex, and IS1 was significant. The data appear to question the “stimulus persistence” model as applied to age differences in the temporal organization of form.
The present investigation was undertaken to examine directly the stimulus-persistence model as applied to age differences in the temporal organization of form. Axelrod (1963) suggested the applicability of the stimulus-persistence model to age differences in temporal resolution and with his colleagues in a later report stated that, “In the senescent nervous system, there may be an increased persistence of the activity evoked by a stimulus . . . ’’ (Axelrod, Thompson, & Cohen, 1968, p. 193). Botwinick (1973) has reviewed a substantial body of literature supportive of the proposition of the increasing persistence of sensory traces with advancing age. For example, critical flicker frequency (CFF) thresholds decline with age (Misiak, 1947), and this decline is particularly prominent
Downloaded by [Deakin University Library] at 21:17 07 November 2015
334
KLINE/BAFFA
for low light: dark ratio presentations (McFarland, Warren, & Karis, 1958). An analogous result for older subjects has been shown for the fusion of pairs of clicks (Weiss, 1963) when young and old adults of ‘‘normal” health were compared (old subjects of “extraordinary” health and young “normals” were not seen to be significantly different on this measure). A similar result has been reported for the fusion of pairs of discrete mild electric shocks (Axelrod, Thompson, & Cohen, 1968).
Recent studies showing greater backward visual masking effects among older adults (Kline & Szafran, 1975; Kline & Birren, 1975), while not directly examining stimulus persistence, can be seen as support for the model. In perhaps the most direct test of the stimulus-persistence model as applied to aging, it has been demonstrated that though spiral and figural aftereffects may be more difficult to establish in the older subject, once established they are more likely to endure (Griew, Fellows, & Howes, 1963; Eisdorfer & Axelrod, 1964). I11 addition to such behavioral evidence for an increase in stimulus persistence with age, Mundy-Castle (1962) observed that EEG aftereffect activity to photic stimulation was more protracted among older subjects. Thus far most of the support for the notion of the increased persistence of stimuli in the older nervous system has been derived from the post hoc application of data from various studies to the model. Direct prospective tests of the model are lacking. However, the appropriate test of this or any other model must as Botwinick (1973) suggests ‘ I . . . be made of predictions the model generates, not post hoc interpretations of existing data” (p. 135). Further, previous investigators have concentrated upon the problems of temporal resolution in the older nervous stystem. The purpose of the present investigation was t o examine prospectively the adequacy of the stimulus persistence model as applied to age differences in the sequential organization of form and to do so in a task in which an age-related increase in stimulus persistence should aid rather than interfere with discriminative perceptual performance. In the present experiment word stimuli to be identified were presented as corresponding stimulus halves
Downloaded by [Deakin University Library] at 21:17 07 November 2015
SEQIJENTIAL INTEGRATION OF FORM
335
with a varied interval between them. As correct stimulus identification is known to decrease as the interval between the halves is increased (Eriksen & Collins, 1967), any age-related increase in the persistence of the stimulus trace should operate to “bridge” this gap, facilitating integration of the word halves. Thus, following the stimulus-persistence model, it was hypothesized that older subjects would do proportionately better on such a task as a n increasing delay was introduced between the presentations of the correponding halves of target stimuli. Specifically, a significant age by interstimulus interval (ISI) interaction was predicted i n which stimulus identification would decline more for young subjects than older ones as IS1 was increased.
METHOD Subjects Two groups, each of 12 unpaid volunteer subjects ( 6 male, 6 female) were employed: an older group, mean age 55.6 years (range 51 to 62) and a younger group, mean age 21.3 years (range 19 to 28). All subjects had at least a high school education; the older group had a mean of 15.6 years formal education, the young group 14.7 years. All subjects employed had a tested visual acuity of 20/25 or better as measured by use of a Snellen chart. Because of procedural screening criteria, the data from 13 older subjects and 4 younger ones were not included in the final tabulation. Apparatus and Materials A Scientific Prototype Model GB 6-channel tachistoscope was used for the presentation of the stimulus materials. Neutral density filters (Kodak) and a Gamma Scientific Photometer Model 700 MA were employed in calibrating the light levels of the tachistoscopic fields. An artificial pupil of 6.5 mm. aperture was placed in the eye piece of the tachistoscope (3.8 cm. from the cornea and therefore approximatdy equivalent to a cornea-adjacent pupil of 2.7 mm.) to control for age decreases in resting pupillary diameter (Birren, Casperson & Botwinick, 1950).
Downloaded by [Deakin University Library] at 21:17 07 November 2015
336
KLINE/BAFFA
A small black cross (0.4' x 0.4') centrally located in the visual field presented for 2 seconds was employed as a fixation point. The fixation point and all test stimuli were presented on a homogeneously illuminated background (2.1 milli-lamberts) occupying a visual angle of 4.2 O vertically by 7.2 O horizontally. Five three-letter words, bag, bug, boy, buy, and bay, were used as test stimuli. Following a technique devised by Eriksen and Collins (1967), these words were constructed of dots. Thus any word could be either presented in its entirety or randomly divided into two corresponding stimulus halves with some varied interval between the halves. The composition of these words was based on a sixdot circle system such that the selective elimination or addition of one or more dots transformed one stimulus word into one of the other four (Fig. 1.). The high inter-item similarity including the random distribution of small extra dots was employed to preclude test stimulus identification from a half-stimulus alone. To minimize the effect of age differences in memory, a limited number of stimuli were used in a recognition format. The test stimuli occupied the central area of the homogeneous visual field (2.8' vertically by 4.6' horizontally).
Procedure
Following testing for visual acuity the subjects were made completely familiar with the test stimuli. The structural difrerences among the five words were carefully pointed out to the subjects using large scale reproductions of each word. This was continued until each subject was satisfied that he could easily discriminate one word from another, which normally took about four to five minutes. The subjects were then seated in a darkened (but not fully dark) laboratory for about 10 minutes and the task was explained to them. A small scale reproduction of each of the five test stimuli was also affixed to the tachistoscope immediately in front of the subject and h e was instructed to refer to it if he had any doubts about any of the stimuli presented during any of the ensuing trials (administered on a forced-choice basis). Following this, the subject was introduced to the task in a practice session beginning with single tachistoscopic presentations of the
Downloaded by [Deakin University Library] at 21:17 07 November 2015
SEQUENTIAL INTEGRATION OF FORM
337
0
0
e
a
a
0
.
a
a
a
a0
a
a
0
a
a 0
0
a
0
a
a a 0
.
a 0
.
a
. 0
0
a Figure 1 . The two bottom dot patterns ~ ~ h csuperimposed n as corresponding stimulus halves result in the top stiinuliis pittcrn in n-llich the word bag can be read,
Downloaded by [Deakin University Library] at 21:17 07 November 2015
338
KLINE/BAFFA
five different words until each in turn was identified. Subsequently, the subject was given 80 practice trials in which the duration of the stimuli was systematically reduced to approximate the stimulus durations employed in the test trials. This practice session took about 15 minutes to complete. In both the practice and test sessions all trials were self-paced. As a dichoptic tachistoscope was used ( i . e . , different visual fields being presented to either the right or left eye), all stimuli were presented to the right eye only. The test trials were given in two segments, the first of which consisted of establishing for each subject, by a method of limits, the minimum 75% -correct recognition duration threshold for randomly presented whole test stimuli (i. e., the minimum duration at which the subject could correctly identify at least 3 of the 4 stimuli from the 2 descending and the 2 ascending series). This threshold was used to establish the duration for presentation of the stimulus halves in the second-segment integration trials; in the second segment both stimulus halves were presented at a duration double that individual’s simple stimulus recognition threshold. The data were not used from any subject whose simple duration r e c ognitkn threshold exceeded 25 m e c .
In the second segment of test trials the corresponding stimulus halves were presented at six different interstimulus intervals (offset-to-onset) ranging from 0 msec. to 150 msec. in 30 msec. increments. One-half of each group of subjects was given the ascending order of IS1 ( i . e . , beginning with 0 msec.) and one-half the descending order (i. e., beginning with 150 msec.). Eleven trials were given at each level of ISI, the first trial of which was considered as practice only. All eleven trials were given at one IS1 before proceding to the next ISI. The number of words correctly identified from the ten test trials presented at each level of IS1 was recorded f o r each subject. Correct stimulus identification is known to decrease with increasing ISI, so to test the prediction of the stimulus-persistence model data were not used from any subject who could not correctly identify at least seven of the ten words presented with an IS1 of 0 msec.
SEQUENTIAL INTEGRATION OF FORM
339
Downloaded by [Deakin University Library] at 21:17 07 November 2015
RESULTS Age differences in duration recognition thresholds were evaluated by t test for independent samples. Age differences on the integration trials were tested utilizing an Age x Sex x Order x IS1 (2 x 2 x 2 x 6) partially repeated measures analysis of variance. It was found that the older group had a significantly higher (p
ISSN: 0361-073X (Print) 1096-4657 (Online) Journal homepage: http://www.tandfonline.com/loi/uear20
Differences in the sequential integration of form as a function of age and interstimulus interval Donald W. Kline & Gary Baffa To cite this article: Donald W. Kline & Gary Baffa (1976) Differences in the sequential integration of form as a function of age and interstimulus interval, Experimental Aging Research, 2:4, 333-342, DOI: 10.1080/03610737608257990 To link to this article: http://dx.doi.org/10.1080/03610737608257990
Published online: 27 Sep 2007.
Submit your article to this journal
Article views: 6
View related articles
Citing articles: 2 View citing articles
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=uear20 Download by: [Deakin University Library]
Date: 07 November 2015, At: 21:17
Downloaded by [Deakin University Library] at 21:17 07 November 2015
DIFFERENCES IN THE SEQUENTIAL INTEGRATION OF FORM AS A FUNCTION OF AGE AND INTERSTIMULUS INTERVAL
DONALD W. KLINE
GARY BAFFA
Psychology Defmtment University of Notre Dame Notre Dame, Jndiana 46556
Andrus Gerontology Center University of Southern California Los Angeles, California 90007
Kline, D. W., & Baffa, G. Differences in the sequential integration of form. Experimental Aging Research, 1976, 2 (4), 333-343. The sequential integration of visual stimuli was studied in 12 young (18-28 years) and 12 old (51-62 years) men and women. Stimuli were constructed as corresponding word halves and inter-stimulos intervals (ISI’s) of varied duration (0-150 msec.) were introduced between the presentations of the two halves. Correct word identification was found to be an inverse function of IS1 for both young and old subjects. T h e number of stimuli correctly identified was significantly lower for the older subjects a t all levels of ISI. None of the interactions of age, sex, and IS1 was significant. The data appear to question the “stimulus persistence” model as applied to age differences in the temporal organization of form.
The present investigation was undertaken to examine directly the stimulus-persistence model as applied to age differences in the temporal organization of form. Axelrod (1963) suggested the applicability of the stimulus-persistence model to age differences in temporal resolution and with his colleagues in a later report stated that, “In the senescent nervous system, there may be an increased persistence of the activity evoked by a stimulus . . . ’’ (Axelrod, Thompson, & Cohen, 1968, p. 193). Botwinick (1973) has reviewed a substantial body of literature supportive of the proposition of the increasing persistence of sensory traces with advancing age. For example, critical flicker frequency (CFF) thresholds decline with age (Misiak, 1947), and this decline is particularly prominent
Downloaded by [Deakin University Library] at 21:17 07 November 2015
334
KLINE/BAFFA
for low light: dark ratio presentations (McFarland, Warren, & Karis, 1958). An analogous result for older subjects has been shown for the fusion of pairs of clicks (Weiss, 1963) when young and old adults of ‘‘normal” health were compared (old subjects of “extraordinary” health and young “normals” were not seen to be significantly different on this measure). A similar result has been reported for the fusion of pairs of discrete mild electric shocks (Axelrod, Thompson, & Cohen, 1968).
Recent studies showing greater backward visual masking effects among older adults (Kline & Szafran, 1975; Kline & Birren, 1975), while not directly examining stimulus persistence, can be seen as support for the model. In perhaps the most direct test of the stimulus-persistence model as applied to aging, it has been demonstrated that though spiral and figural aftereffects may be more difficult to establish in the older subject, once established they are more likely to endure (Griew, Fellows, & Howes, 1963; Eisdorfer & Axelrod, 1964). I11 addition to such behavioral evidence for an increase in stimulus persistence with age, Mundy-Castle (1962) observed that EEG aftereffect activity to photic stimulation was more protracted among older subjects. Thus far most of the support for the notion of the increased persistence of stimuli in the older nervous system has been derived from the post hoc application of data from various studies to the model. Direct prospective tests of the model are lacking. However, the appropriate test of this or any other model must as Botwinick (1973) suggests ‘ I . . . be made of predictions the model generates, not post hoc interpretations of existing data” (p. 135). Further, previous investigators have concentrated upon the problems of temporal resolution in the older nervous stystem. The purpose of the present investigation was t o examine prospectively the adequacy of the stimulus persistence model as applied to age differences in the sequential organization of form and to do so in a task in which an age-related increase in stimulus persistence should aid rather than interfere with discriminative perceptual performance. In the present experiment word stimuli to be identified were presented as corresponding stimulus halves
Downloaded by [Deakin University Library] at 21:17 07 November 2015
SEQIJENTIAL INTEGRATION OF FORM
335
with a varied interval between them. As correct stimulus identification is known to decrease as the interval between the halves is increased (Eriksen & Collins, 1967), any age-related increase in the persistence of the stimulus trace should operate to “bridge” this gap, facilitating integration of the word halves. Thus, following the stimulus-persistence model, it was hypothesized that older subjects would do proportionately better on such a task as a n increasing delay was introduced between the presentations of the correponding halves of target stimuli. Specifically, a significant age by interstimulus interval (ISI) interaction was predicted i n which stimulus identification would decline more for young subjects than older ones as IS1 was increased.
METHOD Subjects Two groups, each of 12 unpaid volunteer subjects ( 6 male, 6 female) were employed: an older group, mean age 55.6 years (range 51 to 62) and a younger group, mean age 21.3 years (range 19 to 28). All subjects had at least a high school education; the older group had a mean of 15.6 years formal education, the young group 14.7 years. All subjects employed had a tested visual acuity of 20/25 or better as measured by use of a Snellen chart. Because of procedural screening criteria, the data from 13 older subjects and 4 younger ones were not included in the final tabulation. Apparatus and Materials A Scientific Prototype Model GB 6-channel tachistoscope was used for the presentation of the stimulus materials. Neutral density filters (Kodak) and a Gamma Scientific Photometer Model 700 MA were employed in calibrating the light levels of the tachistoscopic fields. An artificial pupil of 6.5 mm. aperture was placed in the eye piece of the tachistoscope (3.8 cm. from the cornea and therefore approximatdy equivalent to a cornea-adjacent pupil of 2.7 mm.) to control for age decreases in resting pupillary diameter (Birren, Casperson & Botwinick, 1950).
Downloaded by [Deakin University Library] at 21:17 07 November 2015
336
KLINE/BAFFA
A small black cross (0.4' x 0.4') centrally located in the visual field presented for 2 seconds was employed as a fixation point. The fixation point and all test stimuli were presented on a homogeneously illuminated background (2.1 milli-lamberts) occupying a visual angle of 4.2 O vertically by 7.2 O horizontally. Five three-letter words, bag, bug, boy, buy, and bay, were used as test stimuli. Following a technique devised by Eriksen and Collins (1967), these words were constructed of dots. Thus any word could be either presented in its entirety or randomly divided into two corresponding stimulus halves with some varied interval between the halves. The composition of these words was based on a sixdot circle system such that the selective elimination or addition of one or more dots transformed one stimulus word into one of the other four (Fig. 1.). The high inter-item similarity including the random distribution of small extra dots was employed to preclude test stimulus identification from a half-stimulus alone. To minimize the effect of age differences in memory, a limited number of stimuli were used in a recognition format. The test stimuli occupied the central area of the homogeneous visual field (2.8' vertically by 4.6' horizontally).
Procedure
Following testing for visual acuity the subjects were made completely familiar with the test stimuli. The structural difrerences among the five words were carefully pointed out to the subjects using large scale reproductions of each word. This was continued until each subject was satisfied that he could easily discriminate one word from another, which normally took about four to five minutes. The subjects were then seated in a darkened (but not fully dark) laboratory for about 10 minutes and the task was explained to them. A small scale reproduction of each of the five test stimuli was also affixed to the tachistoscope immediately in front of the subject and h e was instructed to refer to it if he had any doubts about any of the stimuli presented during any of the ensuing trials (administered on a forced-choice basis). Following this, the subject was introduced to the task in a practice session beginning with single tachistoscopic presentations of the
Downloaded by [Deakin University Library] at 21:17 07 November 2015
SEQUENTIAL INTEGRATION OF FORM
337
0
0
e
a
a
0
.
a
a
a
a0
a
a
0
a
a 0
0
a
0
a
a a 0
.
a 0
.
a
. 0
0
a Figure 1 . The two bottom dot patterns ~ ~ h csuperimposed n as corresponding stimulus halves result in the top stiinuliis pittcrn in n-llich the word bag can be read,
Downloaded by [Deakin University Library] at 21:17 07 November 2015
338
KLINE/BAFFA
five different words until each in turn was identified. Subsequently, the subject was given 80 practice trials in which the duration of the stimuli was systematically reduced to approximate the stimulus durations employed in the test trials. This practice session took about 15 minutes to complete. In both the practice and test sessions all trials were self-paced. As a dichoptic tachistoscope was used ( i . e . , different visual fields being presented to either the right or left eye), all stimuli were presented to the right eye only. The test trials were given in two segments, the first of which consisted of establishing for each subject, by a method of limits, the minimum 75% -correct recognition duration threshold for randomly presented whole test stimuli (i. e., the minimum duration at which the subject could correctly identify at least 3 of the 4 stimuli from the 2 descending and the 2 ascending series). This threshold was used to establish the duration for presentation of the stimulus halves in the second-segment integration trials; in the second segment both stimulus halves were presented at a duration double that individual’s simple stimulus recognition threshold. The data were not used from any subject whose simple duration r e c ognitkn threshold exceeded 25 m e c .
In the second segment of test trials the corresponding stimulus halves were presented at six different interstimulus intervals (offset-to-onset) ranging from 0 msec. to 150 msec. in 30 msec. increments. One-half of each group of subjects was given the ascending order of IS1 ( i . e . , beginning with 0 msec.) and one-half the descending order (i. e., beginning with 150 msec.). Eleven trials were given at each level of ISI, the first trial of which was considered as practice only. All eleven trials were given at one IS1 before proceding to the next ISI. The number of words correctly identified from the ten test trials presented at each level of IS1 was recorded f o r each subject. Correct stimulus identification is known to decrease with increasing ISI, so to test the prediction of the stimulus-persistence model data were not used from any subject who could not correctly identify at least seven of the ten words presented with an IS1 of 0 msec.
SEQUENTIAL INTEGRATION OF FORM
339
Downloaded by [Deakin University Library] at 21:17 07 November 2015
RESULTS Age differences in duration recognition thresholds were evaluated by t test for independent samples. Age differences on the integration trials were tested utilizing an Age x Sex x Order x IS1 (2 x 2 x 2 x 6) partially repeated measures analysis of variance. It was found that the older group had a significantly higher (p
