Cognitive Development and the Ageing Process

Selected works of Patrick Rabbitt

Patrick Rabbitt

• Publisher: Routledge
• Year: 2019
• Print ISBN: 9781848723702
• eBook ISBN: 9781317219064
• Topic: Psychology
• Subtopic: Cognitive Psychology & Cognition
• Index: Psychology

In the World Library of Psychologists series, international experts present career-long collections of what they judge to be their most interesting publications – extracts from books, key articles, research findings, practical and theoretical contributions.

Professor Patrick Rabbitt has been a prominent contributor to knowledge of cognitive performance and cognitive ageing for over half a century. He has made a range of significant contributions to geronotological research, from the development of information processing theories in the 1950s and 1960s to a new understanding of decision making and the ageing process in subsequent decades.

This collection of his research articles represents a review of how work in cognitive performance and cognitive ageing has developed in the past 50 years. Whilst the nature of scientific research means that some of the questions posed have since been answered, Rabbitt adds introductory sections to articles which contextualise its place in the subject area and offer a personal view on the evolution of the field.

This book is important because it provides a perspective on the development of cognitive research and the ageing process through the work of an active researcher in the field. It will interest all students and researchers interested in cognitive development and gerontology.

1 An age-decrement in the ability to ignore irrelevant information

P.M.A. Rabbitt¹

Responses made by human beings typically concern circumscribed aspects of their environments, and so are based upon a small fraction of the information simultaneously made available by their sense organs. This selectivity is a condition of survival, since urgent responses may be slowed by the consideration of irrelevant information.

Irrelevant perceptual input can be ignored by selecting between modalities. We may look or listen (Broadbent & Gregory, 1961). Recent experiments have concerned selection between receptors in the same modality, particularly the ears (Broadbent, 1954; Treisman, 1961). Such selection is clearly central, and perhaps most interesting, when discriminations are made between members of a set of complex stimuli which are all of the same kind. The example most frequently investigated has been the “cocktail-party experiment” (Cherry & Taylor, 1953) in which Ss are required to discriminate between competing monologs, usually by repeating one aloud and ignoring the other.

Analogous situations occur in visual search, as when a crowd is scanned for a friend or when a particular word is sought in a dictionary. Techniques for investigating performance in visual search allow the investigator to measure the time taken to ignore irrelevant items (Archer, 1954; Gregg, 1954; Hodge, 1959) and so to quantify results in terms of transmitted information and to make estimates of channel-capacity in this type of task.

Results on young Ss (Rabbitt, 1964a) suggest that selection between relevant and irrelevant items on a visual display may be considered as an example of a more general type of task in which the same response is made to each member of one set of items, while different responses are made to members of other similar sets (Rabbitt, 1959; Broadbent & Gregory, 1962; Pollack 1963). Ss searching a display for relevant items among irrelevant items may be considered to make the responses “relevant₁, relevant₂, relevant₃ . . . relevant_N” to each of N items for which search is conducted and to make the response “irrelevant” to all items in the “irrelevant” set. The response-times of young people are unaffected (Broadbent & Gregory, 1962) or only slightly affected (Rabbitt, 1959; Pollack, 1963) by increases in the number of items in each of the sets between which they discriminate. In contrast, the response-times of old people increase sharply with the number of different items in each set discriminated (Rabbitt, 1964c). If an analogy exists between stimulus categorization tasks and visual search tasks it should follow that the difficulty of discriminating between relevant and irrelevant information increases with age. The experiment described below was made to examine this prediction.

Materials and methods

Materials

Eight packs of 48 visiting card blanks measuring 21/4” × 31/2” were made up, 4 for each of 2 experimental conditions.

Condition 1

Half the cards in each pack were stencilled with the letter A and half with the letter B. These letters were capitals, 5 mm. high, stencilled in india ink. The A or B on each card might be stencilled in any one of nine possible locations (Figure 1.1, iv.) Each pack was made up so that the relevant letter (A or B) appeared, as far as possible, equally often in each of these nine locations. Ss were required to sort these packs into 2 piles, by separating cards marked with A from cards marked with B. The four packs provided displays with four conditions of irrelevant information:

• Pack 1: Only the relevant letter A or B appeared on each card. (Figure 1.1, i).
• Pack 2: Besides the relevant letter A or B, one other letter of the alphabet appeared on each card. These (irrelevant) letters were stencilled in the same size, ink, and type face as the relevant letters. All letters of the alphabet from C—Z were used in turn as irrelevant letters. An irrelevant letter might appear in any of the nine possible locations on the face of a card, and all locations were used in rotation, so that the relative positions of the relevant and irrelevant letters on the face of each card were effectively randomized (Figure 1.1, ii).
• Pack 3: Besides the relevant letter A or B, four irrelevant letters appeared on each card. All letters of the alphabet from C—Z were used in turn as irrelevant letters. All four irrelevant letters on each card were always different. Irrelevant letters were located at random among the nine possible locations on the face of each card (Figure 1.1, iii).
• Pack 4: Besides the relevant letter A or B, eight irrelevant letters appeared on each card. The constraints on selection and location were as described for Packs 2 and 3 (Figure 1.1, iv).

Figure 1.1 Displays used to vary number of irrelevant symbols in Conditions 1 and 2.

Condition 2

Four packs of 48 cards were made up, in which equal numbers of cards in each pack were stencilled with one of the eight relevant letters A, B, C, D, E, F, G, or H. The four conditions of irrelevancy described in Condition 1 were duplicated. Cards in Pack 1 each bore one of these eight relevant letters, but no other symbol; Cards in Pack 2 each bore one relevant and one irrelevant letter; Cards in Pack 3 each bore one relevant and four irrelevant letters, and cards in Pack 4 each bore one relevant and eight irrelevant letters. Irrelevant letters were drawn from the vocabulary of letters I—Z, and were stencilled in the same size, ink, and type face as the relevant letters. Constraints on their selection and location were precisely as described in Condition 1, and the displays were again as illustrated in Figure 1.1, i—iv.

It will be noted that the total vocabulary of relevant and irrelevant letters employed was the same in each condition: that is, the 26 letters of the alphabet A—Z.

Subjects

Two groups of Ss were tested; Group 1 consisted of 11 young people, 3 men and 8 women aged from 17–24 years (mean age 19.0); 5 were women students at a teacher’s training college, 3 were sales personnel at a Cambridge department store, and 3 were undergraduates at the University of Cambridge, England. All volunteered to serve without payment. Group 2 consisted of 11 old Ss aged from 65—74 years (mean age 67.4 years); 2 were men and nine were women. (These Ss were tested at the Section on Aging, N.I.M.H., N.I.H., Besthesda, Md., to which the writer’s thanks are due.) All were members of a retired civil servants’ club in Washington, D.C., U.S.A., and had held very senior positions during their working lives. Most members of this group were university graduates. Although test scores were not available, these Ss were at least the intellectual equals of the young group on any socio-economic criterion.

Procedure

Half the Ss in each group were first tested on Condition 1 and then, on another day, were tested on Condition 2. The remaining Ss reversed the order of practice. Ss were instructed to deal the cards as fast and accurately as possible into the number of piles appropriate to their condition. Packs in each condition were presented in random order until each had been sorted ten times. Ss held the packs face-downwards in their non-dominant hands. Cards were turned and dealt with their dominant hands. This was done to ensure that Ss could not overlap the act of dealing each card with an examination of its successor. Over-all sorting times for each pack were recorded by E with a stopwatch. Each pack was thoroughly shuffled by E before each sorting to randomize the order in which the relevant symbols were encountered.

Results

Errors

Ss in the old group committed errors on 0.012% of their responses when sorting packs in Condition 1 and on 0.014% of their responses in Condition 2. Young Ss committed errors on 0.15% of their responses in Condition 1 and on 0.17% of their responses in Condition 2. Differences between age-groups and Conditions were not significant.

Sorting times

Means and standard deviations of sorting times recorded for each group are given in Table 1.1.

An analysis of variance was carried out on the combined data for the two groups. The details of this analysis are set out in Table 1.2.

Table 1.1 Mean Times (in Seconds) to Sort Packs of 48 Cards into 2 Piles (Condition 1) or 8 Piles (Condition 2) Under Four Conditions of Irrelevant Information.

The main point of interest is whether the sorting-times of the young and old Ss were differently affected by an increase in the number of symbols for which they had to search (Conditions) or by variations in the number of irrelevant symbols on the display (Packs). Table 1.2 shows that the interaction term for Conditions x Age was not significant (P &gt; 0.25). It seems that as the number of relevant symbols was increased from two to eight the increase in sorting time was much the same for old and young people. However, the interaction term for Packs x Age was significant (P < 0.01), which indicates that the sorting times of the old Ss increased more sharply than those of the young with the number of irrelevant letters on a display (Table 1.1). The significant Packs x Conditions interaction (P < 0.001) confirms that increases in the number of irrelevant letters on display affect sorting time more in Condition 2 than in Condition 1. It is particularly interesting that the third order interaction term for Packs x Conditions x Age was also significant (P < 0.01), since this indicates (Table 1.1) that the interaction between Packs and Conditions was more marked for the young group than for the old.

The remaining terms of the analysis listed in Table 1.2 simply confirm that Ss take longer to sort cards into eight piles than into two piles (Conditions P < 0.001), that sorting is slowed by the addition of irrelevant symbols to displays (Packs, P < 0.001) and that the old Ss sort cards more slowly than the young (Age, P < 0.001).

Discussion

The main finding of this experiment is that the speed with which Ss ignore irrelevant symbols on a visual display varies with the number of relevant items for which they are set to search. This result has been discussed in other contexts (Rabbitt, 1962, 1964a). Discussion of the present data turns on the interpretation of the age differences observed in the efficiency with which irrelevant symbols can be ignored.

It is known that young people can simultaneously process several symbols or words when scanning a visual display (Mackworth & Mackworth, 1958) or when reading (Morton, 1964). It has been further suggested that the size of the perceptual sample may vary with the amount of information which a S is required to process (Rabbitt, 1962; 1964a; Morton, 1964). It is possible that when young Ss search for two relevant letters they ignore irrelevant symbols more quickly because they process them in larger groups than when they search for eight relevant letters. The significant interaction-terms obtained for Age vs. Packs and for Age vs. Packs vs. Conditions could therefore be explained on the assumption that old Ss cannot sample letters on a display in large groups in either condition. It is not a new finding that Ss process items on a display in groups. The interest of the present data lies in the suggestion that the size of such groups varies with the number of items for which search is conducted, and with the age of the S.

Table 1.2 Analysis of Variance on Mean Sorting-Times Obtained for Old and Young Groups of Subjects in Both Experimental Conditions.

In the majority of the population visual acuity deteriorates considerably with advancing age (Weiss, 1959). While no S in the present experiment suffered from any serious visual defect, and all who needed correcting lenses wore them, it can be assumed th...

Table of contents

• Cover
• Half Title
• Series Page
• Title Page
• Copyright Page
• Dedication Page
• Table of Contents
• Preface
• List of works reprinted
• 1 An age-decrement in the ability to ignore irrelevant information
• 2 Learning to ignore irrelevant information
• 3 Modulation of selective attention by sequential effects in visual search tasks
• 4 Hand dominance, attention and the choice between responses
• 5 Interference between binary classification judgments and some repetition effects in a serial choice reaction time task
• 6 There are stable individual differences in performance variability, both from moment to moment and from day to day
• 7 Some errors of perceptual analysis in visual search can be detected and corrected
• 8 Processing a display even after you make a response to it. How perceptual errors can be corrected
• 9 Consciousness is slower than you think
• 10 Does it all go together when it goes? The Nineteenth Bartlett Memorial Lecture
• 11 The University of Manchester longitudinal study of cognition in normal healthy old age, 1983 through 2003
• 12 Death, dropout, and longitudinal measurements of cognitive change in old age
• 13 Balance marks cognitive changes in old age because it reflects global brain atrophy and cerebro-arterial blood-flow
• 14 Age-associated losses of brain volume predict longitudinal cognitive declines over 8 to 20 years
• 15 White matter lesions account for all age-related declines in speed but not in intelligence
• 16 Effects of global atrophy, white matter lesions, and cerebral blood flow on age-related changes in speed, memory, intelligence, vocabulary, and frontal function
• Index