Mental image acuity in auditory imagery and perception

Resum

Deciding whether a note is in tune within a musical context is an easy task for some people and very difficult for others. The present doctoral research focuses on the investigation of the formation of auditory mental images of pitch in music. We were interested in finding the cognitive and physiological characteristics underlying the formation of accurate pitch representations in music. More specifically, we have tried to better elucidate the relative influence of the different types of memory systems on perception and imagery. In a series of behavioral experiments we showed that the acuity of pitch representations depended on the combination of several memory factors, such as the tonal relation of the representations with the musical context (long-term memory), the training or repetition of the stimulus (short-term memory) and the assistance of the preceding sensory information presented (Navarro Cebrian y Janata, 2010; JASA). More importantly, we showed how both a bottom-up perception task (in which tones preceding the representation are presented) and a top-down imagery task (in which tones preceding the representation are imagined) are influenced by these different memory systems in a different grade. In this sense, the accuracy of representations formed in an imagery task is more influenced by the experience with the tonal context (long-term memory) while the accuracy of the representations in a bottom-up perception task seems to correlate more with a task in which the representation is equal to a sensory memory trace. In the second stage of my dissertation we have used the ERP technique to study the neural correlates of accurate mental representations (Navarro Cebrian y Janata, 2010; Brain Research). We hypothesized that the quality of an auditory image in a top-down task can be comparable to that of a sensory memory trace, in a bottom-up task, to the point of evoking similar neural responses. In the imagery condition, for example, the amplitude of the N100 component of the auditory evoked potential in response to the target tone was smaller for those listeners who formed more accurate mental images. It was comparable in amplitude to the N100 evoked when all of the notes leading to the target were heard, consistent with a process of habituation of the N100 in the auditory cortex due to the formation of a sequence of mental images. The P3a response, a marker of deviance detection, to mistuned targets was also found in the imagery condition and it was larger for listeners who formed more accurate images. Finally, in a different ERP experiment, we wanted to observe whether the training effects found after some perceptual learning could be generalized to an improvement of the acuity of images. The participants were trained only in a bottom-up (perception) task, and our data showed that there was also an improvement in the top-down (imagery) task in the post-tests with respect to the pretest. There was not only an improvement in performance but also a stronger P3a effect occurring after the last deviated tone in an 'imagined' sequence of tones. These effects were more significant after 48 hours than right after the training session, showing the slow neural changes that develop in the absence of stimulation and are responsible for consolidation in long-term memory.