Object recognition in acquired and developmental prosopagnosia

ABSTRACT

Whether face and object recognition are dissociated in prosopagnosia continues to be debated: a recent review highlighted deficiencies in prior studies regarding the evidence for such a dissociation. Our goal was to study cohorts with acquired and developmental prosopagnosia with a complementary battery of tests of object recognition that address prior limitations, as well as evaluating for residual effects of object expertise. We studied 15 subjects with acquired and 12 subjects with developmental prosopagnosia on three tests: the Old/New Tests, the Cambridge Bicycle Memory Test, and the Expertise-adjusted Test of Car Recognition. Most subjects with developmental prosopagnosia were normal on the Old/New Tests: for acquired prosopagnosia, subjects with occipitotemporal lesions often showed impairments while those with anterior temporal lesions did not. Ten subjects showed a putative classical dissociation between the Cambridge Face and Bicycle Memory Tests, seven of whom had normal reaction times. Both developmental and acquired groups showed reduced car recognition on the expertise-adjusted test, though residual effects of expertise were still evident. Two subjects with developmental prosopagnosia met criteria for normal object recognition across all tests. We conclude that strong evidence for intact object recognition can be found in a few subjects but the majority show deficits, particularly those with the acquired form. Both acquired and developmental forms show residual but reduced object expertise effects.

KEYWORDS:

• Object memory
• face
• object expertise
• cars

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 As an example, our control subjects had a mean score of 37.17 on the Cambridge Bicycle Memory Test, while the linear regression of their scores against age had a slope of 0.08 and an intercept of 35.05. Prosopagnosic subject R-IOT1 was age 49 and his score was 37. His age-predicted score would be 35.05+0.08*49 = 38.90. Hence his deviation from predicted score is 37 – 38.90 = −1.90. To create an age-adjusted score we add this deviation to the control mean performance: −1.90 + 37.17 = 35.27.

2 Here and in the discussion, dissociations on the Cambridge Bicycle Memory Test are considered present only if both the analytic methods we used indicated this.

Additional information

Funding

This work was supported by Canada Research Chairs: [Grant Number 950-228984]; National Eye Institute: [Grant Number F32 EY023479]; Economic and Social Research Council: [Grant Number RES-062-23-2426]; National Science Foundation: [Grant Number 1634098]; Institute of Neurosciences, Mental Health and Addiction: [Grant Number MOP-102567]; Royal Society of New Zealand Marsden Fund: [Grant Number 16-VUW-175].