1. Background Information
 

Author(s)	Robert E. Canestrari, Jr.
Title	Paced and Self-paced Learning in Young and Elderly Adults
Source	Journal of Gerontology, 18, 165-168
Yearequa	1963

 

2. Abstract

This experiment was done to determine whether there was an effect of pacing on the rate of learning by subjects of different ages (young and old adults). Previous studies had shown that deficits in learning by the elderly might be at least partly due to paced schedules. This study utilized paired-associate learning, i.e. learning word pairs, to see how young and old subjects would perform under conditions of 1.5 seconds per trial, 3.0 seconds per trial, and selfpaced where the subject could take as much time as he desired. A significant age difference was found; a significant difference due to speed (pacing), and a significant age X speed interaction were found the younger always performed better than the elderly learners, but the difference in performance was less under the self-paced conditions.

3, Null Hypothesis, sample level and sample size per group.

Ho: Aj=0, where Aj =main effect due to sequence

Ho: Bk=0, where Bk =main effect due to list (,~,B or C)

Ho: G1=0, where G1= main effect due to type of pacing

Ho: Dm=0, where Dm = main effect due to age

Ho: ADjm=0, where ADjm = interaction effect of sequence and age

Ho: BDkm=0, where BDkm = interaction effect of list and age

Ho: GD=0, where GD = interaction effect of age type of pacing

All Hl: effects = 0.

4. Independent and dependent variables.

Independent variables:

Nuisance variables: list and position (sequence presentation)

5. Instrument, briefly comment on its reliability and validity.

The instruments used in this study were learning materials consisting of three paired-associate lists of six pairs of words each. These lists were prepared by 2 lengthy processes which was likely to give equal levels of difficulty to the three lists. The word pairs were selected by choosing 18 stimulus words and then pairing them with response words "...which occurred only once in one thousand presentations of the stimulus words." (pg 165) The lists were presented to subjects similar to the ones used in the study for tests to show equal difficulty. Some adjustments were made, then the lists verified r-tested for equal difficulty. No significant ~ was found on the last test for differences in difficulty between the lists.

6. Experimental Procedure (brief)

The subjects in the study were thought to be similar in SES and educational backgrounds, and in health status because they were recruited at an employment agency where they were looking for work. A WAIS vocabulary subtest showed comparable levels of intelligence for the subjects. Age range for the older group was 60-69 with a mean of 65.4. Range for the younger group was 17-35 with a range of 23.9.

Each subject was given instructions and seated before the Gerbr~nds Memory Drum on which the lists were presented. Under the two rapid-paced conditions, the subjects were told that they had a certain period of time to respond with the response word. Under the self-paced conditions, the subject operated the drum himself and took as much time as he desired to produce the response word. Measurements were taken of the time in seconds per trial under the self-paced conditions. .. m~c;~v~ e was taken of errors made under the various time conditions, and these were also broken down into errors of omission (no response word given) and errors of commission (incorrect response word given).

7. Statistical analysis and conclusion.

A Graeco-Latin Square analysis yielded the following information:

Significant differences found:
 

Age	F=46.16
Speed (pacing)	F=15.92*
age X Speed	F=7.96*

 

non-significant differences found:
 

Position (sequence)	F=1.36
~ist	F=0.12
age X position	F=1.03
Age X List	F= .20

 

Conclusions:

Reject the following null hypotheses:

Ho: G1= 0
Ho: Dm = 0
Ho: GD = 0

There is an effect due to age, and effect due to speed, and an effect due to interaction between age and speed.

Separate analyses of variance provided the following information: Under all conditions of pacing, the younger subjects had fewer errors. The younger subjects used significantly less time on the self-paced test. Differences in omission errors were significant as a function of pacing schedules. There were significant differences in omission errors as a function of pacing schedule when analyzed for each age group separately.

8. If you were the researcher, how would you improve the study? (Be specific)

There is little that I could offer in the way of improvements for this study. The researcher has circumvented one of the most common problems in research with the elderly which is using older subjects whose health is not good, which may have an effect on cognitive performance. There could still be some effects from health conditions such as high blood pressure, which were not obvious, but at least large health effects would not be present since these subjects were all well enough to be looking for employment.

The only question I have is whether there were differences in education and occupational background which could have caused differences in performance. The researcher states that the subjects were questioned about educational level and occupational levels, but makes no reference to summaries of those interviews. Other research has shown differences between age groups on cognitive performance, and a great deal of the variance is due to educational level attained.