Introduction

In the auditory modality, there is growing evidence of cortical plasticity induced not only by profound unilateral deafness, but also by partial sensorineural hearing loss. In the latter case, a severe lesion of the basal turn of the cochlea, where high-frequency tones are encoded, may induce a reorganisation of the primary auditory cortex (Rajan et al., 1993). The cortical area which initially encoded high-frequency tones becomes responsive to tones at the edge of the hearing loss. The lesion-edge frequency is thus over-represented in the auditory cortex. This phenomenon has been observed in many animal species (e.g., in cats: Rajan et al., 1993; in guinea pigs: Robertson and Irvine, 1989; in monkeys: Schwaber et al., 1993). In humans, there is also proof of cortical reorganisation in subjects with severe sensorineural hearing loss: at the cut-off frequency (Fc) of their hearing loss: the amplitude of N1m (a prominent wave reflecting primary auditory cortex activity) is higher there than at lower frequencies (Dietrich et al., 2001). This cortical over-representation may also affect some perceptual abilities. Around the cut-off frequency of a hearing loss, the ability to discriminate frequencies is improved (McDermott et al., 1998; Thai Van et al., 2002). Since no peripheral factors, such as the presence of otoacoustic emissions or the use of loudness clues, have been able to explain this improvement, a central origin is clearly indicated (Thai Van et al., 2003). Such plasticity also has consequences on behaviour. Willott et al. (1994), using a startle modification paradigm in mice, demonstrated an enhanced capacity to modify startle reflex when the animals where hearing tones presented at hearing-loss edge frequencies. This suggests that the salience of tones was perceived differently at these frequencies. The present study investigated whether the cortical reorganisation could also affect reaction time.

Reaction Time (RT) has long been used as a sensory measure. Piéron showed that Simple Reaction Time decreases as a power function of stimulus intensity in various sensory modalities. Extensive data confirming Piéron’s law were provided by Chocholle (1940) for pure tones. In fact, Piéron’s law contains two meaningful parameters, which reflect the effects of two types of factor. The first is sensitivity, which varies with tone frequency in parallel with auditory threshold. Thus, RT will basically be longer in the frequency range concerned by hearing impairment (see review in Scharf, 1978) – although, when tones at different frequencies have the same perceived loudness level, the RTs are identical (Chocholle, 1940; Pfingst et al., 1975; Seitz and Rakerd, 1997). The other parameter is an asymptotic time-constant related to task complexity, to the subject’s ‘willingness to respond’, and more generally to response criterion factors (Bonnet & Dresp, 2001; Link, 1992). For example, Gatehouse and Gordon (1990) demonstrated a decrease in RT for speech stimuli in hearing-aided subjects; this was attributed by the authors to easier perceptual processing rather that to any increase in loudness. One question is what level of processing is mainly involved in these two parameters. The second of the two is clearly related to central processing of the decisional aspects of the task (Bonnet & Dresp, 2001; Link, 1992). The answer is less straightforward concerning the “sensitivity” factor: does it rely on receptor-level or more central processing? As stated above, there are reasons to believe that the enhanced discrimination around the hearing-loss cut-off frequency has a central origin. An absence of any specific effect of this frequency on RT would argue in favour of a peripheral origin for the variation in RT observed with intensity.

Consequently, the overall goal of this study was to investigate whether RT is influenced by the local frequency discrimination improvement situated around the cut-off frequency of steeply sloping hearing loss. Simple RT paradigms were used to test the hypothesis that RT would not be shortened at the frequency at which frequency discrimination performances were the best (bDLF) if they had a peripheral origin. If an alteration in RT was observed at the bDLF, this would indicate that central reorganisation can modify the behaviour of steeply-sloping hearing-impaired listeners. The consequences of auditory rehabilitation on RT performance were also investigated, first in general, and then specifically at the bDLF.