Effects of learning on somatosensory decision-making and experiences

Abstract

Medically unexplained symptoms (MUS), where patients experience disabling physical symptoms in the absence of medical pathology, are very common in medicine but poorly understood and difficult to manage. Recent models, reviewed in Chapter 1, identified MUS as somatic misperceptions and suggested that training to reduce somatosensory misperception more generally might result in decreased symptom reporting. Researchers have used an experimental paradigm called the somatosensory signal detection task (SSDT) to study somatosensory misperception in the laboratory. In this task, participants often report illusory touch sensations ("false alarms") when vibrations are absent, which are believed to be an experimental analogue of MUS. This thesis presents a series of studies examining the potential for training to change false alarm rates on the SSDT and other perceptual tasks, with a view to informing MUS treatment. The training was based on operant conditioning and its effects on SSDT performance were investigated in four studies. Each study had two conditions (control vs. training) and three phases (baseline vs. manipulation vs. follow-up). Studies 1 and 2 used a within-subjects design. Studies 4 and 5 used a between-subjects design and addressed the methodological limitations of Studies 1 and 2. Using money as a reinforcer, Studies 1 and 4 randomly rewarded 50% of hits and punished 50% of misses in the manipulation phase of the experimental condition, with the aim of increasing the false alarm rate for individuals initially low in false alarms. Studies 2 and 5 randomly rewarded 50% of correct rejections and punished 50% of false alarms in the manipulation phase of the experimental condition, with the aim of decreasing the false alarm rate for individuals initially high in false alarms. Training in Studies 1 and 4 significantly increased false alarm and hit rates, made response criterion more liberal, but did not change sensitivity between the conditions. Training in Studies 2 and 5 significantly decreased false alarm and hit rates, made response criterion more stringent, but did not change sensitivity between the conditions. A new voice-hearing task was developed in Study 3 taking into account the limitations of existing paradigms and was used to examine transfer of training in Studies 4 and 5. Effects of training did not transfer to a second tactile task called the spontaneous sensation test but there was some evidence for transfer on the voice-hearing task, although the effects fell short of statistical significance. Voice false alarms correlated positively with the SSDT false alarm rate in the baseline phase, suggesting that a common mechanism underlies illusory perception in the different senses. Study 6 was guided by the idea that successive SSDT trials would be sequentially dependent due to the activation of somatosensory schema underlying SSDT responses. All four SSDT studies were thus examined to see if responses in current trials were affected by stimuli and responses in previous trials. Data indicated that current yes responses were more likely to be preceded by another yes response or the presence of vibration on N-1 trials, whereas the light had no such effect. Sequential dependencies were approximately normally distributed and some were affected by the SSDT training. A significant positive relationship was found between somatization and sequential effect of N-1 light on misses, although no such association was found for SSDT false alarms. My findings add to previous efforts to train participants on the SSDT. The conditioning paradigm can potentially be used independently or in combination with other procedures aiming to change somatic experiences. The voice-hearing task is a useful paradigm to study psychotic phenomena and cross-modal transfer. In sum, the studies strongly suggest that illusory perceptual experiences are trainable, which might have important implications for the treatment of MUS.

Date of Award	1 Aug 2016
Original language	English
Awarding Institution	The University of Manchester
Supervisor	Ellen Poliakoff (Supervisor) & Richard Brown (Supervisor)