Search

Acquiring non-native speech contrasts can be difficult. A seminal study by Logan, Lively and Pisoni (1991) established the effectiveness of phonetic training for improving non-native speech perception: Japanese learners of English were trained to perceive /r/-/l/ using minimal pairs over 15 training sessions. A pre/post-test design established learning and generalisation. In a follow-up study, Lively, Logan and Pisoni (1993) presented further evidence which suggested that talker variability in training stimuli was crucial in leading to greater generalisation. These findings have been very influential and “high variability phonetic training” is now a standard methodology in the field. However, while the general benefit of phonetic training is well replicated, the evidence for an advantage of high over lower variability training remains mixed. In a large-scale replication of the original studies using updated statistical analyses we test whether learners generalise more after phonetic training using multiple talkers over a single talker. We find that listeners learn in both multiple and single talker conditions. However, in training, we find no difference in how well listeners learn for high vs low variability training. When comparing generalisation to novel talkers after training in relation to pre-training accuracy, we find ambiguous evidence for a high-variability benefit over low-variability training: This means that if a high-variability benefit exists, the effect is much smaller than originally thought, such that it cannot be detected in our sample of 166 listeners.

Contains additional details on the methods, as well as a table of results from the Representational Similarity Analysis of brain activation data.

From only a single spoken word, listeners can form a wealth of first impressions of a person’s character traits and personality based on their voice. However, due to the substantial within-person variability in voices, these trait judgements are likely to be highly stimulus-dependent for unfamiliar voices: the same person may sound very trustworthy in one recording but less trustworthy in another. How trait judgements may change when listeners are familiar with a voice is unclear: Are listeners who are familiar with the voices as susceptible to the effects of within-person variability? Does the semantic knowledge listeners have about a familiar person influence their judgements? In the current study, we aimed to empirically test the effect of familiarity on listeners’ trait judgements from variable voices, by tracking how first impressions may differ from second (or “lasting”) impressions. For this purpose, we conducted a series of 3 experiments in which we contrasted trait judgments for listeners who were familiar with a set of voices – either through lab-based training or through watching a TV show – with listeners who were unfamiliar with the voices. We predicted that familiarity would reduce variability in trait judgements for variable voice recordings from the same identity (cf. Mileva, Kramer & Burton, 2019 for faces). However, across the 3 studies and two types of measures to assess variability, we found no compelling evidence to suggest that trait impressions were systematically affected by familiarity.

When we hear a voice, we instantly form rich impressions of the person it belongs to – whether we are familiar with this voice or whether we are hearing it for the first time. Despite the rich impressions we can form of both familiar and unfamiliar voices, current models of voice processing primarily focus on familiar voice identity perception only and do not explicitly account for the processing of unfamiliar voices. Where unfamiliar identity processing is described, it tends to be in the context of specific identity perception tasks, such that the extant literature is largely built on a distinction between familiar voice recognition and unfamiliar voice discrimination. We argue that the current focus of the literature is too narrow in its strong emphasis on identity-specific perception, and does not adequately reflect person perception from voices beyond experimental tasks. Here, we propose a broader, unified account of person perception from both familiar and unfamiliar voices. We suggest that listeners routinely perceive all person characteristics from voices via common recognition processes, based on representations – be those of a specific identity, speaker sex, accent, or a perceived personality trait. While explicit discrimination processes may still be used to disambiguate percepts, they are likely to play a smaller role in perception in naturalistic settings. We offer discussions of how this representation-centred person perception from voices may work, in terms of the nature of representations, their specificity and interactions of different kinds of representation.

High variability training has been shown to benefit the learning of new face identities. In two experiments, we investigated whether this is also the case for voice identity learning. In Experiment 1, we contrasted high variability training sets including stimuli extracted from a number of different recording sessions, speaking environments and speaking style with low variability stimulus sets that only included a single speaking style (read speech) extracted from one recording session (see Ritchie & Burton, 2017 for faces). In Experiment 2, variability was manipulated in terms of the number of unique items as opposed to number of unique speaking contexts/styles. Here, we contrasted the high variability training sets used in Experiment 1 with low variability training sets that included the same breadth of contexts/styles, but fewer unique items; instead, individual items were repeated (see Murphy, Ipser, Gaigg & Cook, 2015 for faces). For both studies, listeners were trained on 4 voice identities (2 identities through high variability training and 2 identities through low variability training) and were tested on an old/new recognition task using read sentences. We found no high variability training advantage in Experiment 1 – instead we found a disadvantage. In Experiment 2, we only found weak evidence for a high variability advantage, an effect that can be explained by stimulus-specific effects. Thus, we do not find conclusive evidence that high variability training aids the learning of novel voice identities. We discuss these findings in the context of mechanisms thought to underpin advantages for high variability training.

Supplemental material, QJE-STD-18-292.R1-Supplementary_Material for Breaking voice identity perception: Expressive voices are more confusable for listeners by Nadine Lavan, Luke FK Burston, Paayal Ladwa, Siobhan E Merriman, Sarah Knight and Carolyn McGettigan in Quarterly Journal of Experimental Psychology

Human voices are extremely variable: The same person can sound very different depending on whether they are speaking, laughing, shouting or whispering. In order to successfully recognise someone from their voice, a listener needs to be able to generalise across these different vocal signals ('telling people together'). However, in most studies of voice identity processing to date, the substantial within-person variability has been eliminated through the use of highly controlled stimuli, thus focussing on how we tell people apart. We argue that this obscures our understanding of voice identity processing by controlling away an essential feature of vocal stimuli that may include diagnostic information. In this paper, we propose that we need to extend the focus of voice identity research to account for both 'telling people together' as well as 'telling people apart'. That is, we must account for whether, and to what extent, listeners can overcome within-person variability to obtain a stable percept of person identity from vocal cues. To do this, our theoretical and methodological frameworks need to be adjusted to explicitly include the study of within-person variability.

Within-person variability is a striking feature of human voices: our voices sound different depending on the context (laughing vs. talking to a child vs. giving a speech). When perceiving speaker identities, listeners therefore need to not only "tell people apart" (perceiving exemplars from two different speakers as separate identities) but also "tell people together" (perceiving different exemplars from the same speaker as a single identity). In the current study, we investigated how such natural within-person variability affects voice identity perception. Using voices from a popular TV show, listeners, who were either familiar or unfamiliar with the show, sorted naturally-varying voice clips from 2 speakers into clusters to represent perceived identities. Across three independent participant samples, unfamiliar listeners perceived more identities than familiar listeners and frequently mistook exemplars from the same speaker to be different identities. These findings point towards a selective failure in "telling people together". Our study highlights within-person variability as a key feature of voices that has striking effects on (unfamiliar) voice identity perception. Our findings not only open up a new line of enquiry in the field of voice perception but also call for a re-evaluation of theoretical models to account for natural variability during identity perception.