A novel behavioural approach to detecting tinnitus in the guinea pig

Abstract

Tinnitus, the perception of sound in the absence of an external stimulus, is a particularly challenging condition to demonstrate in animals. In any animal model, objective confirmation of tinnitus is essential before we can study the neural changes that produce it. A gap detection method, based on prepulse inhibition of the whole-body startle reflex, is often used as a behavioural test for tinnitus in rodents. However, in the guinea pig the whole-body startle reflex is subject to rapid habituation and hence is not an ideal behavioural measure. By contrast, in this species the Preyer or pinna reflex is a very reliable indicator of the startle response and is much less subject to habituation. We have developed a novel adaptation of the gap detection paradigm, which uses the Preyer reflex to measure the startle response, rather than whole-body movement. Using this method, we have demonstrated changes in gap detection, in guinea pigs where tinnitus had been induced by the administration of a high dose of salicylate. Our data indicate that the Preyer reflex gap detection method is a reliable test for tinnitus in guinea pigs.

Fig. 1

Schematic of the gap detection test, adapted from Turner and Parrish (2008). A startling pulse in continuous background noise elicits a startle reaction (A). When a gap (50 ms) is presented before the startling pulse, this reduces the amplitude of the startle response (B). When an animal is experiencing tinnitus, it will have difficulty detecting the gap, as this will be partially filled in by the tinnitus, and will show less inhibition of the startle response (C).

Fig. 2

A photograph showing the position of the reflective markers when fixed to the pinnae. Arrows indicate the direction of movement of the pinna in response to a startling auditory stimulus.

Fig. 3

Representative raw traces of ‘no gap’ startle stimulus presentation (n = 10) overlaid for the Preyer reflex (A) and the WBS (B), taken from a single trial for one guinea pig. Dotted lines indicate the window of analysis for WBS and peak-to-peak measurement for Preyer reflex.

Fig. 4

PPI of the WBS and Preyer reflexes. Mean PPI of each reflex at a given background noise frequency is indicated by the solid horizontal line contained within each box; boxes indicate 95% confidence intervals; whiskers indicate the full range of values obtained across all GPs (n = 12) for each condition. PPI of WBS was significantly higher than the Preyer reflex in the BBN condition (**P < 0.01).

Fig. 5

The effects of sodium salicylate on PPI of the Preyer reflex (A) and the WBS (B). Mean (± SEM) PPI values for all GPs (n = 4) are shown for each background frequency noise condition at 2 h, 5 h, and 72 h post-salicylate administration. For the Preyer reflex, significant reductions in PPI were seen at 2 h in the BBN condition and at 5 h in the 8–10 kHz condition (**P < 0.01). For the WBS, no significant changes were observed at any time point.

Fig. 6

Changes in amplitude of the Preyer (A) and WBS (B) responses. Mean (± SEM) values are shown for all guinea pigs (n = 4) at each background noise frequency before (baseline) and 2 h after salicylate administration. Significant increases in response amplitude were detected at 16–18 kHz (**P < 0.01) for the Preyer reflex, and 4–6 and 8–10 kHz (*P < 0.05; ***P < 0.0001) for the WBS.