Anglia Ruskin Research Online (ARRO)
Browse
Jones_2015.PDF (497.31 kB)

Learning Faces: Similar Comparator Faces Do Not Improve Performance

Download (497.31 kB)
journal contribution
posted on 2023-07-26, 14:13 authored by William J. Matthews, Scott P. Jones, Dominic M. Dwyer, Michael B. Lewis
Recent evidence indicates that comparison of two similar faces can aid subsequent discrimination between them. However, the fact that discrimination between two faces is facilitated by comparing them directly does not demonstrate that comparison produces a general improvement in the processing of faces. It remains an open question whether the opportunity to compare a “target” face to similar faces can facilitate the discrimination of the exposed target face from other nonexposed faces. In Experiment 1, selection of a target face from an array of novel foils was not facilitated by intermixed exposure to the target and comparators of the same sex. Experiment 2 also found no advantage for similar comparators (morphed towards the target) over unmorphed same sex comparators, or over repeated target exposure alone. But all repeated exposure conditions produced better performance than a single brief presentation of the target. Experiment 3 again demonstrated that repeated exposure produced equivalent learning in same sex and different sex comparator conditions, and also showed that increasing the number of same sex or different sex comparators failed to improve identification. In all three experiments, exposure to a target alongside similar comparators failed to support selection of the target from novel test stimuli to a greater degree than exposure alongside dissimilar comparators or repeated target exposure alone. The current results suggest that the facilitatory effects of comparison during exposure may be limited to improving discrimination between exposed stimuli, and thus our results do not support the idea that providing the opportunity for comparison is a practical means for improving face identification.

History

Refereed

  • Yes

Volume

10

Issue number

1

Page range

e0116707

Publication title

PLOS ONE

ISSN

1932-6203

Publisher

Public Library of Science

File version

  • Published version

Language

  • eng

Legacy posted date

2017-11-29

Legacy creation date

2017-11-29

Legacy Faculty/School/Department

ARCHIVED Faculty of Science & Technology (until September 2018)

Usage metrics

    ARU Outputs

    Categories

    No categories selected

    Licence

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC