Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 36 (6), 1182-95

Why Are Some People's Names Easier to Learn Than Others? The Effects of Face Similarity on Memory for Face-Name Associations

Affiliations

Why Are Some People's Names Easier to Learn Than Others? The Effects of Face Similarity on Memory for Face-Name Associations

Peter C Pantelis et al. Mem Cognit.

Abstract

Using synthetic faces that varied along four perceptual dimensions (Wilson, Loffler, & Wilkinson, 2002), we examined the effects of face similarity on memory for face-name associations. The nature of these stimuli allowed us to go beyond the categorical similarity manipulations used in previous verbal associative memory studies to trace out the parametric relation between similarity and various performance measures. In Experiment 1, we found that recall performance diminished as a function of how many studied faces were in the vicinity of the cue face in similarity space. Also, incorrect recalls were more likely to come from nearby positions in face space. Experiments 2 and 3, respectively, demonstrated analogous effects with a set of more distinguishable, photorealistic faces, and in an associative recognition task. These results highlight the similarity between associative recall and associative recognition, and between face-name association and other domains of associative memory.

Figures

Figure 1
Figure 1
Sample face–name pairs from (A) Experiments 1 and 3 and (B) Experiment 2.
Figure 2
Figure 2
Face stimuli used in Experiment 1, plotted in their 4-dimensional MDS-derived face space. The 4th dimension is indicated by the brightness of the face; negative coordinates in this dimension are associated with darker colors.
Figure 3
Figure 3
Illustration of the procedures used during the introductory same–different task (A), study phases (B), and test phases (C) of Experiment 1.
Figure 4
Figure 4
The “neighborhood” effect. Because of scarcity of data for faces with seven neighbors, faces with six and seven neighbors are collapsed into one bin. Error bars represent ±1 standard error of the mean.
Figure 5
Figure 5
Intralist intrusions. (A) The conditional probability of making an intralist intrusion, as a function of the Euclidean distance between the face corresponding to the recalled name and the cue face. (B) The corresponding response latencies for these intrusions. Error bars represent ±1 standard error of the mean.
Figure 6
Figure 6
The “neighborhood” effect in Experiment 2. Low = 0–1 neighbors; medium = 2–3; high = 4–5. Error bars represent ±1 standard error of the mean.
Figure 7
Figure 7
Conditional probability of intralist intrusion in Experiment 2 as a function of the Euclidean distance between the face corresponding to the recalled name and the cue face. Error bars represent ±1 standard error of the mean.
Figure 8
Figure 8
Average confidence judgment (A) and reaction time (B) as functions of Euclidean distance between the probe face and the face belonging to the probe name. In panel A, confidence judgments above the dotted line were defined as a response of “correct pairing,” and judgments below the dotted line represented a response of “incorrect pairing.” Error bars represent ±1 standard error of the mean.
Figure 9
Figure 9
Final cued recall. (A) The conditional probability of recalling a name as a function of the Euclidean distance between the cue face and the corresponding face of the name (a correct recall has a distance of zero). (B) The corresponding reaction times of these recalls. Error bars represent ±1 standard error of the mean.

Similar articles

See all similar articles

Cited by 8 articles

See all "Cited by" articles

Publication types

LinkOut - more resources

Feedback