Atypical Minds: the Cognitive Science of Difference and Potentialities
Cognitio 2015
Young researchers conference in cognitive science
Montréal, June 8th, 9th and 10th 2015
Persons in Our Heads: Voice Hearing, Social Cognition and the Brain
Nikola Andonovski.
One of the most puzzling features of “voice hearing” experiences (“hearing voices” in the absence of external stimuli) is the frequent personalization of the voices. Voice hearers often experience voices not only as possessing specific sensory qualities, but also as having strikingly vivid ‘personalities’. Indeed, phenomenological studies indicate that up to 70% of voice hearers experience voices that have distinct personal identities. The voices are often ‘attributed’ to existing people, but also to supernatural agents such as God, angels, or the Devil. Given the frequency of some voice hearing experiences, it is perhaps unsurprising that almost two thirds of voice hearers indicate that they develop ‘relationships’ with their voices. Intriguingly, these relationships seem to share many properties with ‘normal’ interpersonal relationships. Review studies report that the ways the hearers relate to their voices often mirror the ways they relate to people of flesh and blood. For example, individuals who perceive themselves to be of low social rank (inferior) also feel inferior in relation to their voices and behave accordingly. Despite these numbers, and the growing body of literature exploring the relevance of the concept of ‘relating’ for these experiences, the neuro-cognitive theories of voice hearing have largely ignored the interpersonal issues. Thus, the two major family of theories – self-monitoring theories and “spontaneous activation” theories – have had very little to say about the personalization of voices.
In this paper, I sketch a neurocognitive theory of the mental mechanisms that support such personalization. I propose that voice hearers experience their voices as personalized because they rely on ‘normal’ mechanisms for voice recognition and social cognition. According to the proposal, aberrant and persistent activations of auditory representations can, via automatic mechanisms of voice processing, lead to creation of mental models of the voices’ ‘characters’ and “personality traits”. The theory has two components. The first involves the mechanisms of voice perception and recognition that, the argument goes, are activated in both ‘normal’ audition and experiences of voice hearing. According to the extensive empirical literature, listeners utilize sensory cues in order to form representations of speaker identity, as well as to differentiate between distinct familiar and unfamiliar speakers. In addition, studies show that listeners are extremely good at determining or inferring the physical, psychological and social characteristics of speakers solely from their voices. They also use sensory cues to form basic personality impressions of the speakers. If these mechanisms of voice processing are automatically activated during auditory experiences of voices, it is to be expected that they be also activated during hallucinatory experiences triggered by aberrant activations of the sensory cortices. The consequences can be very significant. Persistent spontaneous activation of the same sensory areas – and consequently hearing the ‘same’ voices repeatedly - can lead to the formation of representations of speaker identity: the voices will be ‘recognized’ as belonging to specific ‘agents’ that possess certain physical, psychological and social characteristics.
The second component appeals to recent work in social neuroscience, proposing that the personalization of the voices is not exhausted by the representation of their basic characteristics. According to relevant studies, people do not only attribute individual beliefs and desires to others; they also internally represent their enduring personality traits. In addition, some have argued that the brain automatically constructs internal models of specific social actors for both reasoning and social interaction. These results suggest promising solutions to the puzzle of voice-voice hearer ‘relationships’. I suggest that the persistent hearing of ‘recognized’ voices could lead to creation of mental models of these voices or their “personality traits”. For voices that do not belong to actual people, the automatic construction of the models will utilize two types of data: (the inferences drawn from) the sensory cues and the semantic contents of the experiences (what the voices say). While the created models will, undoubtedly, be comparatively simple (next to the models of real social actors), this is precisely what we would expect given the character of the ‘relationships’ hearers develop with their voices. Namely, the vast majority of the hearers experience voices that have one dominant physical/psychological characteristic.
The wager of this essay is that voice hearers experience their voices as personalized because they rely on ‘normal’ mechanisms for voice recognition and social cognition. To put it provocatively, voice hearers have persons living “in their heads” because, in a sense, we all do.