Knight, R.A.; (2011) Integration of sensory cues by the head direction system. Doctoral thesis , UCL (University College London).

Preview

PDF 1335838.pdf Download (33MB)

Abstract

Head direction (HD) cells fire as a function of the animal’s heading direction, with each cell responding to a specific head orientation. This thesis describes the findings from single cell recordings of HD cells in rats. In particular, the thesis focuses on the manner in which these cells integrate landmark and path integration cues. The first experimental chapter aimed to assess the degree to which HD cell firing is influenced by the geometry of an environment. The findings from this experiment suggest that the reliability of the path integration signal affects the degree to which HD cells integrate geometric cues. As the stability of the path integration signal increased (rat was oriented) the weighting of geometric cues decreased. Conversely, when the stability of the path integration signal decreased (rat was disoriented) the weighting of geometric cues increased. This finding, that the influence of a cue is inversely proportional to the variability of that cue, supports a Bayesian account of cue integration. The second experiment therefore directly tested a Bayesian model of cue integration. HD cells and behaviour were simultaneously recorded during a conflict between path integration signals and landmark information. Behavioural findings did not support a Bayesian model, but recordings of HD cells did show evidence of cue weighting based on cue reliability. Interestingly, the reliability of the cues was not expressed in the cells’ firing rates or tuning widths and therefore the reliability coding could occur before the signal is sent to the HD system. The final experiment had three main objectives: To establish when HD cells switch from primarily using landmark information to mostly using path integration information. The second aim was to reveal whether HD cells in different areas of the HD circuit respond to these two sensory cues in a similar manner. The third aim was to provide an insight into how activity in the HD network propagates from one preferred firing direction to another. HD cell firing demonstrated cue integration across the system with no differences in brain region. At small conflicts, HD cell firing suggested landmark dominance, whilst larger conflicts demonstrated a greater weighting of path integration. Preliminary findings also suggest that activity can sweep from one preferred firing direction to the next.

Download activity - last month

Export as XMLJSONCSV

Download activity - last 12 months

Export as CSVXMLJSON

Downloads by country - last 12 months

Export as XMLJSONCSV

Archive Staff Only

View Item