Frequency selectivity measured both psychophysically and physiologically

Wells, Tobias T. (2014) Frequency selectivity measured both psychophysically and physiologically. PhD thesis, University of Nottingham.

PDF - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (9MB) | Preview


The ability to resolve the individual frequency components of a complex sound is known as frequency selectivity. The auditory system seems to act as a series of overlapping band-pass filters or auditory filters (AF), the width of which describe frequency selectivity. It is a fundamental and extensively studied property of the auditory system, yet its neural basis is not fully understood. This is due, in part, to the fact that two distinct approaches are taken to explore it; psychophysics and physiology, the results of which are difficult to reconcile. AF are measured in quite different ways in the two sub-fields.

Psychophysics measures the ability of the system as a whole by using masking paradigms to measure the bandwidth of AF from behavioural data. A preferred method in human psychophysics is notched noise (NN) masking using forward masking with fixed signal level, since it does not suffer from confounds like suppression and off-frequency listening. In physiology bandwidth can be measured for single cells, or a population of cells, at various levels of the auditory system, and is traditionally done by observing the number of action potentials elicited in response to pure tone stimuli. The auditory system is known to be highly non-linear and so comparing the results of such vastly different approaches is problematic. Also, psychophysics measures the detection threshold of 'signal' sounds; how this compares to mean spike rates used in physiology is not clear.

Attempts have been made in the past to apply the same method in animals to measure both psychophysical and physiological bandwidths, with varying degree of success. No successful attempt has been made to use an up-to-date method used in human psychophysics. In this thesis I take a step towards comparing psychophysical and physiological results by 1) developing a novel method that allows forward masked NN bandwidths to be measured behaviourally in the ferret, and 2) applying the same psychophysical paradigm to measuring bandwidth in guinea pig inferior colliculus (IC) and primary auditory cortex (A1) neurons. In addition a signal detection theory (SDT) approach is used on the physiological data to make results more comparable to psychophysical ones.

Results from the behavioural method show that it can be used to successfully measure both forward and simultaneously masked NN bandwidths in the same animal, and that these measurements are in close agreement with one another and with bandwidths measured using previous methods. Results from the guinea pig physiology study show that bandwidths measured from IC neurons using the psychophysical NN paradigm are narrower than pure tone estimates of bandwidth, in the same neurons. However, the NN estimates are in close agreement with auditory peripheral and perceptual bandwidths, a finding which differs substantially from previous studies. Unexpectedly, however, bandwidths estimated from A1 neurons using masking show much finer tuning at high frequencies than seen further down the auditory system. This tuning is not only narrower than pure tone tuning in these neurons, but also finer than psychophysically measured estimates, which represent the auditory system as a whole. However, this may be related to the greater non-linearity of cortical neurons compared with those in the midbrain and lower.

This work demonstrates that it is possible to reconcile different measurements of tuning in the auditory system by using appropriate methods. It also highlights the complex nature of auditory neurons and how care must be taken when measuring frequency selectivity using different approaches. In addition it provides a method for measuring auditory bandwidths psychophysically and physiologically in the same animal, allowing a direct comparison between the two; a vital step in investigating the neural basis of perceptual frequency selectivity.

Item Type: Thesis (University of Nottingham only) (PhD)
Supervisors: Sumner, C.J.
Mason, R.
Keywords: Auditory perception, Psychoacoustics, Perception in animals
Subjects: Q Science > QP Physiology
Faculties/Schools: UK Campuses > Faculty of Medicine and Health Sciences > School of Life Sciences
Item ID: 14312
Depositing User: EP, Services
Date Deposited: 02 Dec 2014 13:08
Last Modified: 16 Dec 2017 17:25

Actions (Archive Staff Only)

Edit View Edit View