User Identity Protection in Automatic Emotion Recognition through Disguised Speech

Ambient Assisted Living (AAL) technologies are being developed which could assist elderly people to live healthy and active lives. These technologies have been used to monitor people’s daily exercises, consumption of calories and sleep patterns, and to provide coaching interventions to foster positive behaviour. Speech and audio processing can be used to complement such AAL technologies to inform interventions for healthy ageing by analyzing speech data captured in the user’s home. However, collection of data in home settings presents challenges. One of the most pressing challenges concerns how to manage privacy and data protection. To address this issue, we proposed a low cost system for recording disguised speech signals which can protect user identity by using pitch shifting. The disguised speech so recorded can then be used for training machine learning models for affective behaviour monitoring. Affective behaviour could provide an indicator of the onset of mental health issues such as depression and cognitive impairment, and help develop clinical tools for automatically detecting and monitoring disease progression. In this article, acoustic features extracted from the non-disguised and disguised speech are evaluated in an affect recognition task using six different machine learning classification methods. The results of transfer learning from non-disguised to disguised speech are also demonstrated. We have identified sets of acoustic features which are not affected by the pitch shifting algorithm and also evaluated them in affect recognition. We found that, while the non-disguised speech signal gives the best Unweighted Average Recall (UAR) of 80.01%, the disguised speech signal only causes a slight degradation of performance, reaching 76.29%. The transfer learning from non-disguised to disguised speech results in a reduction of UAR (65.13%). However, feature selection improves the UAR (68.32%). This approach forms part of a large project which includes health and wellbeing monitoring and coaching.

Investigating the relationship between Psychopathy, Fear Conditioning, and Facial Affect Recognition

<p>Psychopathic offenders are often considered to be untreatable, especially dangerous, and at very high risk of reoffending. Psychopathy has generated considerable research interest. Despite this interest, our understanding of psychopathy is relatively poor, with ongoing debate regarding how best to define psychopathy, and a lack of clarity regarding how psychopathy develops. Etiological theories of psychopathy posit deficits in recognising and responding to others’ emotions, and an attenuated experience of fear as crucial mechanisms in the development of psychopathy. The aims of this thesis are to investigate the pattern of psychopathic traits present within an inmate sample, and to investigate the relationship between these psychopathic traits and performance on two tasks related to etiological theories of psychopathy: facial affect recognition and fear conditioning. Part One of this thesis addresses the first aim, investigating the presentation of psychopathy in the current sample. The relationship between psychopathic traits in the present sample was largely consistent with previous research. A Principal Components Analysis identified two factors of psychopathic traits: a Bold/ Fearlessness factor which measures an absence of fear and anxiety and the presence of self-assurance, and a Mean/ Disinhibited factor which measures the presence of externalising and disinhibited behaviour, alongside aggression and the use of other people for one’s own gain. These findings are discussed in relation to common conceptualisations and operationalisations of psychopathy. Part Two of this thesis uses the measurement of psychopathy from Part One to investigate performance on a facial affect recognition task and a fear conditioning task. The Violence Inhibition Mechanism theory suggests that psychopaths should show impairments on facial affect recognition tasks, particularly in the recognition of fearful and sad facial expressions. However, in the current research psychopathy was unrelated to affect recognition, across all emotional expressions. When criminal offenders were compared to a student sample, the offenders showed poorer affect recognition than the students. These results suggest that there may be an effect of antisociality on affect recognition, but no effect of psychopathy. Low fear theories of psychopathy suggest that psychopaths should be impaired at learning conditioned fear associations. However, the present study found no evidence of psychopathy-related deficits in fear conditioning. Rather, higher psychopathy was related to better fear conditioning, with higher scores on the Mean/ Disinhibited factor predicting better discrimination between the conditioned and neutral stimuli. Taken together, these findings suggest that psychopathy was not related to deficits in either affect recognition or fear conditioning. These findings are inconsistent with etiological theories of psychopathy, and question common assumptions about the deficits which characterise psychopathy.</p>

Investigating the relationship between Psychopathy, Fear Conditioning, and Facial Affect Recognition

<p>Psychopathic offenders are often considered to be untreatable, especially dangerous, and at very high risk of reoffending. Psychopathy has generated considerable research interest. Despite this interest, our understanding of psychopathy is relatively poor, with ongoing debate regarding how best to define psychopathy, and a lack of clarity regarding how psychopathy develops. Etiological theories of psychopathy posit deficits in recognising and responding to others’ emotions, and an attenuated experience of fear as crucial mechanisms in the development of psychopathy. The aims of this thesis are to investigate the pattern of psychopathic traits present within an inmate sample, and to investigate the relationship between these psychopathic traits and performance on two tasks related to etiological theories of psychopathy: facial affect recognition and fear conditioning. Part One of this thesis addresses the first aim, investigating the presentation of psychopathy in the current sample. The relationship between psychopathic traits in the present sample was largely consistent with previous research. A Principal Components Analysis identified two factors of psychopathic traits: a Bold/ Fearlessness factor which measures an absence of fear and anxiety and the presence of self-assurance, and a Mean/ Disinhibited factor which measures the presence of externalising and disinhibited behaviour, alongside aggression and the use of other people for one’s own gain. These findings are discussed in relation to common conceptualisations and operationalisations of psychopathy. Part Two of this thesis uses the measurement of psychopathy from Part One to investigate performance on a facial affect recognition task and a fear conditioning task. The Violence Inhibition Mechanism theory suggests that psychopaths should show impairments on facial affect recognition tasks, particularly in the recognition of fearful and sad facial expressions. However, in the current research psychopathy was unrelated to affect recognition, across all emotional expressions. When criminal offenders were compared to a student sample, the offenders showed poorer affect recognition than the students. These results suggest that there may be an effect of antisociality on affect recognition, but no effect of psychopathy. Low fear theories of psychopathy suggest that psychopaths should be impaired at learning conditioned fear associations. However, the present study found no evidence of psychopathy-related deficits in fear conditioning. Rather, higher psychopathy was related to better fear conditioning, with higher scores on the Mean/ Disinhibited factor predicting better discrimination between the conditioned and neutral stimuli. Taken together, these findings suggest that psychopathy was not related to deficits in either affect recognition or fear conditioning. These findings are inconsistent with etiological theories of psychopathy, and question common assumptions about the deficits which characterise psychopathy.</p>