The Glasgow Voice Memory Test: Assessing the ability to memorize and recognize unfamiliar voices

Abstract

One thousand one hundred and twenty subjects as well as a developmental phonagnosic subject (KH) along with age-matched controls performed the Glasgow Voice Memory Test, which assesses the ability to encode and immediately recognize, through an old/new judgment, both unfamiliar voices (delivered as vowels, making language requirements minimal) and bell sounds. The inclusion of non-vocal stimuli allows the detection of significant dissociations between the two categories (vocal vs. non-vocal stimuli). The distributions of accuracy and sensitivity scores (d') reflected a wide range of individual differences in voice recognition performance in the population. As expected, KH showed a dissociation between the recognition of voices and bell sounds, her performance being significantly poorer than matched controls for voices but not for bells. By providing normative data of a large sample and by testing a developmental phonagnosic subject, we demonstrated that the Glasgow Voice Memory Test, available online and accessible from all over the world, can be a valid screening tool (~5 min) for a preliminary detection of potential cases of phonagnosia and of "super recognizers" for voices.

Deficits in Voice-Identity Processing: Acquired and Developmental Phonagnosia

“…However, AS, an apperceptive developmental phonagnosic, was significantly impaired on both the voice and the bell recognition task (see Figure 4). Aglieri et al (2016). AS's voice and bell recognition were statistically different from controls; SP's performance on both tasks was comparable to controls (AS: voice: p = 0.04, bell: p = 0.002 SP: voice: p = 0.14, bell: p = 0.40).…”

“…Recently, attempts have been made to overcome such language constraints with the launch of the Glasgow Voice Memory Test (GVMT; Aglieri et al, 2016), a brief test, which examines voice-identity processing. Specifically, in this test, listeners are exposed to a series of unfamiliar vocal identities uttering a single vowel and a series of unique bell sounds.…”

“…Moreover, the bell condition permits for an assessment of the specificity of a voice-identity processing deficit. Aglieri et al (2016) noted that the GVMT was sensitive in characterising KH as phonagnosic. Specifically, KH's scores for voice, rather than bell, recognition were significantly poorer than controls.…”

“…The first study on acquired phonagnosia was published in 1976 (Assal et al, 1976) and on developmental phonagnosia in 2009 (Garrido et al, 2009). (ii) Testing of voice-recognition deficits is relatively difficult, as standard tests are not readily available and are often languagedependent (but see Aglieri et al, 2016). (iii) Cases of phonagnosia are rare, although this perceived rarity may be more related to a low self-awareness, rather than a low prevalence rate, of voice-identity processing disorders (Roswandowitz et al, 2014).…”

How many voices did you hear? Natural variability disrupts identity perception from unfamiliar voices

“…Across sets and listener groups, substantial individual differences are furthermore apparent: For familiar listeners, this may reflect the inclusion of participants with a varied duration of exposure (between one and five seasons with variable amounts of time having passed since watching the show) and engagement with the show. Furthermore, these differences could also reflect more general individual differences in voice identity processing (see Aglieri et al ., ).…”