Object Location Memory

(Authors: Kathleen A. Flannery & Marianna Eddy,  Editor:   Maureen McCarthy)

Introduction

Investigators have documented sex differences in cognitive abilities across a variety of tasks. Men are typically better at mental rotation tasks (Halpern, 1992; Linn & Petersen, 1985), and women demonstrate better verbal fluency (Halpern, 1992; Spreen & Strauss, 1991). In general, research suggests that males outperform females on tasks involving spatial ability, whereas females outperform males on tasks involving verbal ability (Collaer & Hines, 1995). However, Silverman and Eals (1992) developed a task that measured object location memory and reported that females outperformed males on a paper and pencil version of this task. James and Kimura (1997) also reported that women outperformed males on the original version of this task when the location of pairs was exchanged, but not when the locations of objects were shifted to sites previously unoccupied by another object. Levy, Astur and Frick (2005) found that women outperformed men when novel objects were substituted and women had to identify the new objects. Barnfield (1999) found that adult data replicated these earlier findings, but children did not show sex differences for this task.

Given that this task has generated a great deal of theoretical speculation ranging from sociobiological (James & Kimura, 1997; Silverman & Eals, 1992) to hormonal explanations (Barnfield, 1999), we created a computer version of object location task developed by Silverman and Eals (1992). The computer version of this task translates the paper and pencil version for testing object location memory into an internet-based experiment that includes multiple object location memory trials that allow us to examine sex differences.

Procedure

The experiment begins with a description of the object location memory task that the participants will complete. Participants learn that they will study an array of objects for a designated period of time; the array of objects will then disappear; and then the array of objects will reappear, but some of the objects will have exchanged positions. In order to designate whether an object is in an exchanged location, the participant can use the mouse to mark the objects that have moved. Practice trials precede participation in the experiment. The practice array consists of 10 objects.

Following the practice trials, participants must examine an array of 27 objects. Participants have up to 5 trials to solve the object location memory task.

Design

The between-subjects variable of interest is gender. The dependent variables are the number of hits, misses, false alarms, and correct rejections. A hit is recorded when the participant correctly identifies one of the 14 objects that moved location. A miss is recorded when the participant fails to identify one of the 14 objects that moved. A false alarm is recorded when the participant falsely indicates that one of 13 stationary objects moved location. A correct rejection is recorded when the participant correctly leaves one of the 13 stationary objects unchecked.

The output includes a composite discrimination index d' (Banks, 1980). This index involves using the percentage of false alarms, or false alarm rate, for each trial and the percentage of hits, or hit rate, for each trial (Claremont Graduate University WISE Project, n.d.). The formula for the Discrimination Index appears in the Appendix.

The dependent measures are reported for each of 5 trials so that the full experimental design is a 2 (Gender) x 5 (trials) mixed design.

Data Download Format & Data Analyses

Data are downloadable in three formats (XML, Excel spreadsheet format, and comma delimited for statistical software packages). Figure 1 shows an excerpt from a sample Excel spreadsheet. The first five columns provide basic data (participant ID number, class ID, gender, age, and date). The subsequent 25 columns contain data from the 5 trials.

Sample data appears below:

Sample data image from the object location memory Data experiment
Figure 1

Trial data are labeled as T1-T5. The number of hits or misses are recorded using the following classification:
TRIAL 1-5 RH - Hit (Correctly identifies object has moved)
TRIAL 1-5 RM - Correct Rejection (Correctly indicates the object did not move)
TRIAL 1-5 WH - False Alarm (Incorrectly identifies object has moved)
TRIAL 1-5 WM - Miss (Incorrectly identifies that object did not move)

The maximum number of hits is 14. In other words, 14 objects moved locations. Thirteen of the items do not move location, for a total of 27 items.

Because there are multiple dependent variables, the user must select the variables of interest. A comparison of the discrimination index (d') across the five trials may offer an interesting analysis. The advantage of using the d' measure is that it is a composite measure of performance. Each of the trials includes a discrimination index value labeled ADI. An ANOVA could compare the discrimination index or performance on trials across males and females using a split plot ANOVA.

In addition to examining composite performance, it may be worthwhile to compare males and females on specific performance variables--hits, misses, false alarms, and correct rejections.

Applications/Extensions

Because earlier researchers (James & Kimura, 1997; Silverman & Eals, 1992) found mixed results for tasks of spatial reasoning on the basis of gender, it might be interesting to examine gender differences for performance in this virtual replication of a classic object memory activity. Recent research suggests that women may outperform men on object memory tasks, and it would be interesting to see if this result holds true for this activity. Similarly, a discussion of the results may include applications of this research. For example, if women use landmarks as cues for directions, then superior performance by women on this task would support this explanation of a difference between men and women. Given Barnfield's (1999) recent study of this task with respect to age, it may also be useful to examine performance on the basis of age and gender.

References

Banks, W. P. (1970). Signal detection theory and human memory. Psychological Bulletin, 74, 
	81-99.  

Barnfield, A. M. (1999) Development of sex differences in spatial memory. Perceptual and Motor 
	Skills, 89, 339-350. 
 
Claremont Graduate University WISE Project (n.d.). Signal detection theory: Gaussian model. 
	Retrieved July 27, 2006, from http://wise.cgu.edu/sdt/index.html

Collaer, M. L., & Hines, M. (1995). Human behavioral sex differences: A role for gonadal 
	hormones during early development? Psychological Bulletin, 118, 55-107. 
 
Halpern, D. F. (1992). Sex differences in cognitive abilities (2nd ed.). Hillsdale, NJ: Erlbaum.  

James, T. W., & Kimura, D. (1997). Sex differences in remembering the locations of objects in an 
	array: Location-shifts versus location-exchanges. Evolution and Human Behavior, 18, 
	155-163.  

Levy, L. J., Astur, R. S., & Frick, K. M. (2005). Men and women differ in object memory but 
	not performance of a virtual radial maze. Behavioral Neuroscience, 119, 853-862.

Linn, M. C., & Petersen, A. C. (1985). Emergence and characterization of sex differences in 
	spatial ability: A meta-analysis. Child Development, 56, 1479-1498. 

Pezdek, K., Whetstone, T., Reynolds, K., Askari, N., & Dougherty, T. (1989). Memory for 
	real-world scenes: The role of consistency with schema expectation. Journal of 
	Experimental Psychology: Learning Memory and Cognition, 15, 587-595.  

Silverman, I., & Eals, M. (1992).  Sex differences in spatial abilities: Evolutionary 
	theory and data. In J. H. Barkow, L. Cosmides & J. Tooby (Eds.), The adapted mind: 
	Evolutionary psychology and the generation of culture (pp.533-549). New York: Oxford 
	University Press.  

Spreen, O., & Strauss, E. (1991). A compendium of neuropsychological tests: Administration, 
	norms, and commentary. New York: Oxford University Press. 

					

Appendix

Calculation of DI - Discrimination index. Hit rate is the proportion of correctly identified moved objects. In order to calculate hit rate for use with this equation, simply obtain the number of Hits and calculate a proportion out of a maximum number of 14. False alarm rate would be the proportion of incorrect identifications that an object had moved. Calculation of this proportion entails using the reported number of false alarms and calculation of a proportion using a maximum number of 13 false alarms.

object location memory equation

In this study there were 14 exchanged objects, therefore hit rate would be the number correctly identified hits divided by 14. There were another 13 unexchanged items in this study. False alarm rate would be the number of items incorrectly identified as exchanged, divided by 13.

A similar measure that one can compute is the composite measure used by Silverman and Eals (1992). This measure, which they call the object location memory score, results from adding the number of hits and correct rejections together and subtracting the number of false alarms and misses from this total.