User:EmmaGibson3/sandbox

Alternative Project idea: Voice Perception.

There is a page on face perception that is really detailed, which talks about the cognitive neuroscience behind it as well as basic principles, but there is not a page for voice perception which is important in aspects of neuropsychology such as Psychoacoustics.

Plan for Project:

Definition of Voice Perception; Picture of model (Belin et al., 2004) and how this relates to face processing; Similarities between voice and face processing (history); Voice Recognition; Encoding Voices; Pathway (neurological); What can happen when areas are damaged; Different disorders that can occur; Gender Differences; References; Further Reading

= Voice Perception =

Voice perception is the ability to recognise familiar voices and distinguish them from unfamiliar voices. Voices disclose socially relevant information about a person especially when focusing on the tone and expressions. They can carry physical characteristics such as gender, age, size and the ability to identify the voice. They can also divulge into mood and affective state as well as more subtle cues such as perceived attractiveness or dominance.

‘Auditory Face’
The voice contains more information than just auditory stimuli, such as speech, it carries important ‘invariant’ or ‘static’ features of a person’s voice. This includes things such as timbre, which focuses on physical features such as age and gender of the speaker; it also carries ‘dynamic’ information such as a person’s accent or their individual laugh. Affective information can be drawn too, identifying the speakers affective state-how they are feeling when speaking.

Voices have often been referred to as an auditory face, inferring that face processing and voice processing are very similar. In 1986, Bruce and Young developed a model of face processing indicating the several steps needed when doing so. In later years, Belin created a replica model for voice processing, with similar steps. These included an initial low- level analysis which occurs in the sub-cortical nuclei and other major regions within the auditory cortex. Voices are then processed in a specific stage called ‘structural encoding’. From this step, three types of information are extracted and further processed through different pathways within the model. The first: an analysis pathway of speech information; this includes passing through important auditory areas such as the Superior temporal sulcus, inferior prefrontal regions and the premotor cortex; commonly in the left hemisphere of the brain. A second pathway is found for the analysis of vocal affective information, commonly found in the temporo-medial regions, anterior insula, amygdala and inferior prefrontal regions; predominantly in the right hemisphere. The last pathway involves the voice recognition units located in the right anterior Superior temporal sulcus; this is for the analysis of vocal identity and will be activated when an individual can recognise the voice. Interactions between the pathways are suggested to occur during normal voice processing in a healthy individual.

The model does not impose that all aspects of face and voice processing are identical, the physical properties of seeing or hearing a stimuli are different, but how they are processed within the brain are extremely similar.

Voices are special
Voices are the most important sound within the auditory environment, and it is possible that it is the most common sound to hear. They are needed for protection and correctly perceiving what a voice is saying can help with survival. Vocal information are special to the brain as pathways indicated in the voice processing model have shown to be connected within one another; it is suggested that vocal stimuli can only be accessed through the structural encoding stage, suggesting there are face specific regions within the brain.

There are indications that there are voice selective cerebral processes, for example, fMRI studies have established the activation of voice-selective populations. These areas have been located within the Temporal Voice Area (TVA) which is located along the middle and anterior part of the Superior temporal gyrus bilaterally. A greater Blood-oxygen-level dependent imaging (BOLD) signal was shown in response to vocal sounds in comparison to non-vocal sounds and scrambled voices. It has also been found that this response is also detected in non-speech sounds in the Temporal voice areas in the right hemisphere.

Voice selective areas have not only been found in adults but also in 7 month old infants; Near-infrared spectroscopy has indicated responses to voice-selectivity within the children, indicating that voices are processed before speech is developed. This may be due to the amount of auditory stimuli a child is introduced to at such a young age. An increase in the Haemodynamic response was found in the superior temporal cortex in both hemispheres when presenting a vocal stimuli in comparison to a non-vocal. Electrophysiological aspects have been applied such as Event-related potential (ERPs) observing the difference between vocal and non-vocal sounds. Vocal sounds had a latency at around 320ms, indicating a voice sensitive response. However, a more recent study indicated a latency of around 200ms which is closely linked to the face processing evidence.

Neuroanatomy
The main area associated with voice processing is the Superior temporal sulcus. Belin & Zatorre in 2003 found that by using an adaptive paradigm, they found the right anterior Superior temporal sulcus is sensitive to changes in identify of the speaker but not lexical phonetic change. Speech perception has shown activation in similar areas such as the superior temporal cortex, heschl's gyrus as well as the middle anterior Superior temporal sulcus.

Animals
Voice processing is not just specific to humans, animal research has produce data to present the findings that non-human primates, such as Rhesus macaque are able to recognise other individuals by their voice. The anterior superior temporal plane areas have responded to conspecific vocalisations in comparison to other sounds, suggesting this is a voice-selective area within the monkey. The right anterior superior temporal plane area is seen as being specialised for processing voice information in humans. Moreover, the anterior ‘what’ processing stream was identified within the monkey brains when listening to vocal sounds. This pathway is important as it is processing specific calls made by monkeys, enabling them to recognise the different voices within their troop. Contrastingly to human brains, the responses are much more ventral, lying close to the superior temporal sulcus. This distinction between the two could have occurred due to evolutionary changes to the structure to the brain over time.

Research has also been conducted on non-primate animals such as dogs. It has been discovered, through Magnetic resonance imaging studies, that dogs have specialised areas within their brain for processing voices. It was collected that voice sensitivity was localised within the Perisylvian cortex; this included areas such as the slyvian gyri, sylvian fissure, ectosylvian gyri, ectosylvian sulcus, with development into the suprasylvian sulcus. These were similar areas in comparison to human brains, with both sets of regions moving towards the temporal pole.