[ CogSci Summaries home | UP | email ]

Salthouse, T. A., & Mitchell, D. R. D. (1990). Effects of age and naturally occurring experience on spatial visualization performance. Developmental Psychology, 26, 845-854.

Author of the summary: David Zach Hambrick, 1998, gt8781a@prism.gatech.edu

Salthouse & Mitchell (1990)


The disuse hypothesis states that age-related declines in cognitive functioning are attributable to disuse of cognitive abilities. This view has been articulated many times. One approach to investigating this hypothesis involves age-comparisons among people presumed to differ in amount of a particular type of experience. Salthouse explains that studies falling within this category can be classified as broad or narrow in terms of assessment of experience or cognition, or both. The problem with broad assessments of experience and cognition is that "the relations between experience and cognition are likely to be quite weak when those constructs are assessed in very general terms" (p. 845).


Another approach involves investigating age-cognition relations among individuals with a particular type of experience. For example, Glanzer and Glaser examined age-relations with cognitive and perceptual in a sample of military pilots. The largest age-relations were found for a task most similar to piloting, at least in terms of face validity. More recently, Salthouse et al. found that architects and non-architects showed comparable rates of age-related decline in performance on tests of spatial visualization ability. However, it is unclear whether the architects in this study were representative of all architects.


The goal of this study was to recruit participants from the general population, and then to see whether self-report estimates of experience with spatial tasks moderated relations between age and spatial visualization ability. Experience questions asked about cumulative and recent experience performing spatial tasks, and the battery of cognitive tests included tests of spatial visualization ability, reasoning, and perceptual processing speed. Including the reasoning and perceptual speed tests provided a check on the validity of the spatial tasks (i.e., higher correlations of experience with spatial than with reasoning or speed measures). Mediation and moderation hypotheses were examined. That is, does age have effects on spatial visualization ability after controlling for experience. Or does the effect of age on spatial visualization performance depend on level of experience, such that people with more experience show less decline?




Experience measures were entered into a principal-components analysis. Factors representing cumulative experience, self-rating of spatial ability, and recent experience were found. Cumulative experience increased with age, while recent experience decreased. The remaining factors were interpreted as reflecting experience with specific types of spatial tasks (e.g., directions, clothes, etc.). Experience estimates for individuals presumed to have a particular type of experience (e.g., architects) were compared with the rest of the sample. The subsample had more experience with particular types of activities, such as considering how an object would look from different perspectives. This provides some evidence for the validity of the experience measures.


Age relations with spatial visualization performance were examined after control of 1) experience measures (components 2, 4, and 8), 2) perceptual speed, and 3) experience measures and perceptual speed. Controlling for experience resulted in a modest attenuation (40%) of the age-related variance in spatial visualization performance. However, restriction of the analyses to Components 4 and 8 (Component 2 was a self-rating) resulted in a 13% reduction. Age x experience interactions predicting inductive reasoning and perceptual speed were non-significant. Some of the age x experience interactions predicting spatial visualization performance were significant. However, inspection of regression lines relating age and spatial visualization performance for three ranges on the component measures showed that the age functions were nearly parallel. Thus, as Salthouse and Mitchell explain, "This suggests that the age effects are similar throughout the range of component values and implies that it is not the case that the magnitude of the age effects is attenuated among individuals with the greatest amoung of experience or self-assessed ability" (p. 851).


Comments and Questions


Salthouse and Mitchell conclude that "The major conclusion implied from our findings is that many of the age-related effects on spatial visualization observed in this study, and presumably in other studies, seem to be relatively independent of the amount of relevant experience the individuals have received" (p. 852). This assumption is, however, predicated on a number of assumptions. The first assumption concerns the validity of the self-reported estimates of experience. The factor analysis revealed coherent and interpretable factors, and individuals in occupations that involve spatial visualization reported more experience of certain types that the other participants. Salthouse and Mitchell therefore had some confidence in their ratings. Whether the questions were appropriate is another matter. The second assumption concerns the range of experience estimates. Was the experience range adequate—enough people with high and enough people with low experience? In fact, there was a considerable range.


The final, and perhaps most serious objection, concerns the possibility that requirements of the spatial visualization tasks and activities included in the experience questionnaire are different. That is, "An objection could be raised that the requirements of these activities are not sufficiently similar to what the examinee must do in the Paper Folding and Surface Development tests used in the assessment of spatial visualization ability to expect substantial relations between experience and spatial visualization performance" (p. 853). If fact, the correlations between the experience factors and spatial visualization scores were quite low. Salthouse and Mitchell acknowledge this as a legitimate concern, but also point out the following (p. 853):


It is interesting to consider the implications for the disuse hypothesis of the difficulty of finding activities relevant to the abilities observed to decrease with increased age: If there are no activities that provide appropriate experience, then neither the concepts of use nor disuse may be very meaningful with respect to the maintenance of decline of spatial abilities across the life span.


What do these results say about individual differences in age-cognition relations? They seem to suggest that they are not attributable to age-differences in experience. And "the seemingly inescapable conclusion from this body of evidence is that many of the age-related effects on measures of relatively basic abilities are largely independent of the amount of relevant experience" (p. 853).


Salthouse, Babcock, Skovronek, Mitchell, & Palmon (1990)


This project consisted of three studies. In Study 1, participants between 20 and 70 were administered several spatial visualization tasks. Then, their performance was scaled in terms of SD units based on performance of a reference sample of 50 young adults (i.e., observed score – average young / SD young). Expected age-related declines in the measures of spatial visualization ability were found.


Study 2 compared older individuals presumed to differ in amount of experience related to spatial visualization tasks. The participants’ scores were expressed in SD units based on the young sample, and the major finding was that the age differences may have been less dramatic for the architects. The architects outperformed the non-architects in terms of difference from the young adults.


The goal of Study 3 was to clarify the results of Study 2 by examining two interpretations of the older architects’ superior performance. The preserved differentiation view holds that "differences between the two groups in their 60s are merely continuations of differences that existed when the individuals were young adults" (p. 132). By contrast, the differential preservation view attributes the difference to the benefits of experience, that is, "to the extensive amount of experience with spatial visualization activities on the part of the architects" (p. 132).


The results can be summarized in brief: age-related declines in spatial visualization performance were comparable for the architects and non-architects. As I have noted, one possibility is that types of experience change, although the gross measures of experience would not have captured this. As Salthouse et al. state, "One possible interpretation is a shift in the pattern of activities within the same occupation so that as the architects become older they spend less time actually using their spatial visualization abilities and that it is this lack of recent exercise that is responsible for the observed age-related declines in spatial visualization performance" (p. 134). However, in this study, all participants (with the exception of 3) rated the measured abilities as highly relevant. Moreover, although there was a negative correlation between age and estimates of experience (which could account for the declines), the negative age-spatial relations persisted even after controlling for amount of experience.


Comments and Questions


A number of issues are discussed in interpreting the results. First, was the type of spatial visualization ability measured relevant to the types of spatial visualization architects use in their occupational activities? Were the measures valid, in other words? Second, have there been historical changes in selection criteria for architecture programs? More specifically, do architecture programs place more emphasis on spatial abilities than in the past. This could be one explanation of why the younger architects outperformed the older architects. Salthouse et al. argue that this interpretation seems unlikely given that similar age trends were found for unselected adults. In summary, "If it is accepted that the present sample of architects had considerable experience using relevant spatial visualization abilities, then the results of the current studies seem to imply that increased age is associated with lower levels of spatial visualization ability even among individuals who are using these abilities extensively in their occupation" (p. 135).


Back to the Cognitive Science Summaries homepage
Cognitive Science Summaries Webmaster:
JimDavies (jim@jimdavies.org)
Last modified: Tue Mar 7 15:30:27 EST 2000