Habituation and sensitization in rat auditory evoked potentials: a single-trial analysis with wavelet denoising

https://doi.org/10.1016/S0167-8760(01)00157-XGet rights and content

Abstract

In this work, systematic changes of single-trial auditory evoked potentials elicited in rats were studied. Single-trial evoked potentials were obtained with the help of wavelet denoising, a very recently proposed method that has already been shown to be useful in the analysis of scalp human evoked potentials. For the evoked components in the 13–24-ms range (i.e. P13, N18, P20 and N24), it was possible to identify slow exponential decreases in the peak amplitudes, most likely related to a slow habituation process, while for N18, an initial increase in amplitude was also found. On the contrary, the slower components (N38 and N52) habituated within a few trials, and we therefore propose that they are related to a different functional process. The outcomes of the present study show that wavelet denoising is a useful technique for analyzing evoked potentials in rats at the single-trial level. In fact, in the present study it was possible to obtain more information than the one described in previous related works. This allows the study of other forms of learning processes in rats with the aid of evoked potentials. Finally, the outcomes of this study may have some relevance for the comparison of human and rat evoked potentials.

Introduction

Sensory evoked potentials represent the activity of large groups of neurons or neural ensembles closely synchronized with stimulus events (Swick et al., 1994). They usually have a low amplitude in comparison with the background EEG, and therefore, ensemble averaging of EEG segments time-locked to the stimuli has been used to visualize the evoked responses. The averaged evoked potentials consist of a series of positive and negative waves, which are often identified by their latency from stimulus presentation. Averaging accomplishes a reduction in the number of data, as well as an increase in the signal-to-noise ratio. However, ensemble averaging assumes that each response contains a background EEG acting as an additive stationary noise, contaminating a time-locked and invariant response independent of the ongoing EEG. These assumptions are, in a strict sense, not valid (see e.g. Basar, 1980). In fact, one fundamental characteristic of evoked potentials is their change from trial to trial; this information being lost in the averaged evoked potentials.

Amplitude and latency trial-to-trial variability in the components of an evoked potential can be either systematic or unsystematic. Systematic changes in response strength to repeated stimulation is a key property of all organisms with a nervous system. New stimuli first elicit an orchestrated response, named the orientation response (Sokolov, 1960). It consists of changes in a large number of autonomic variables, such as the skin conductance reaction, phasic heart rate changes, and it also involves a reaction at the cerebral cortex. Habituation is, in a wide sense, defined as a response diminishment as a function of stimulus repetition and it is a constant finding in almost any behavioral response. It can be best described by an exponential decay if a single novel stimulus is repeated regularly (Sokolov, 1960). Its rate of decay will depend on physical properties of the stimulus, on the interstimulus interval and on the psychological impact of the stimulus, such as its relevance. Also, increases in the responses during the first stimuli have been described, these being related to a sensitization process (Thompson and Spencer, 1966, Groves and Thompson, 1970). Moreover, these authors proposed that response to repeated stimulation is the interaction of sensitization and habituation.

In order to study these systematic changes over time, recording sessions were subdivided into (consecutive) sets of averages of a few trials, namely sub-ensemble averages. This approach is successful only if changes within trials are much slower than the number of trials included in each sub-average. Another approach to study single-trial changes is to repeat a sequence (block) of trials, each sequence elicited after a certain recovery time. Then, single-trial changes (within a block) can be better visualized from an ‘average block’ of trials. This second method, which we will call ‘block-averaging’, assumes that changes between blocks are negligible, a requirement that is not always fulfilled. Moreover, both sub-ensemble and block-averaging require long recording sessions, and thus general arousal changes are likely to occur. Despite these drawbacks, there are quite a number of studies which demonstrate within-session changes of midlatency or late EP components in humans (Davis et al., 1966, Ritter et al., 1968, Fruhstorfer et al., 1970, Groves and Thompson, 1970, Calloway, 1973, Polich, 1989, Lew and Polich, 1993, Polich and McIsaac, 1994, Boutros and Belger, 1999, Carrillo-de-la-Pena and Garcia-Larrea, 1999) and in rats (Hall, 1968, Herr et al., 1994, Miyazato et al., 1994, Miyazato et al., 1996, Miyazato et al., 1999, Boutros et al., 1997). While investigating habituation of the vertex auditory evoked potentials, sub-ensemble averaging in rats was recently applied (de Bruin et al., 2001). With sub-ensembles of only five trials, an indication was found that habituation occurred for two components: the P17 showed a rapid (between trials 6–10 and 11–15) and significant decrease in the amplitude and the N22 also showed a slower but significant decrease. Other changes within the first 100 trials remained unnoticed. As we will show in the present study, more information will be obtained by means of single-trial analysis. Although several methods have been proposed for the analysis of single trials of the evoked responses (see a review in Lopes da Silva, 1993), up to now, none of these attempts has been successful, at least at a level that they could be applied to different type of EPs and be implemented in clinical settings (Quian Quiroga, 2000).

A method based on the wavelet transform, namely wavelet denoising, has recently been described and successfully applied to the analysis of scalp human EPs (Quian Quiroga, 2000). It allowed the visualization of single-trial responses and a further quantitative analysis of their variability, something quite difficult to achieve from the original data. The objective of this study was to apply wavelet denoising to the study of auditory single trial EPs elicited in rats in the data set from de Bruin et al. (2001). In particular we will show changes in the evoked responses during the recording session, these being related to habituation and sensitization processes.

Section snippets

Subjects and data recording

Auditory evoked potentials (AEPs) were obtained from 13 adult male albino rats of different genetic origin (four random-bred Wistar, four APO-SUS, two APO-UNSUS, and three WAG/Rij rats). Rats were 6 months old and weighed 395 (±17) g. They were maintained on a 12-h light–dark cycle with white lights on at 07:00 h. They had ad lib access to standard food and tap water. Animals were treated in accordance with the ‘Principles of Laboratory Animal Care’ (NIH) and institutional guidelines.

Under

Results

The average EPs of a typical rat, with and without denoising, are shown in the uppermost plots of Fig. 1 and the first 10 single trials corresponding to this average are shown in Fig. 2. The gray traces are the original signals and the black traces the denoised ones. The denoised signals follow the high frequencies of the original signal in approximately the first 30 ms after stimulation. For the later times, where only slow EP components are expected, denoising acts as a low-pass filter.

We

Habituation of the evoked responses

Wavelet denoising led to the identification and analysis of several peaks elicited after auditory stimulation in rats, a larger number of peaks than the ones described in previous reports (Boutros et al., 1997, Miyazato et al., 1999, de Bruin et al., 2001). Next, systematic amplitude decreases of the EP components with increasing trial number were observed, these being related to habituation processes. The amplitude changes were best described by a slow exponential decay for the P13, N18, P20

Acknowledgements

We would like to thank CISP for supporting a visit by Dr Rodrigo Quian Quiroga to Nijmegen University, to Dr Natasja de Bruin for giving us access to her data files and to Dr Marijtje Jongsma for assisting in the curve-fitting procedures.

References (42)

  • G.S. Lew et al.

    P300, habituation, and response mode

    Physiol. Behav.

    (1993)
  • H. Miyazato et al.

    Midlatency auditory evoked potentials and the startle response in the rat

    Neuroscience

    (1996)
  • H. Miyazato et al.

    Midlatency auditory evoked potentials in the rat: effects of interventions that modulate arousal

    Brain Res. Bull.

    (1999)
  • M. Molnar

    On the origin of the P3 event-related potential component

    Int. J. Psychophysiol.

    (1994)
  • J. Polich et al.

    Comparison of auditory P300 habituation from active and passive conditions

    Int. J. Psychophysiol.

    (1994)
  • R. Quian Quiroga et al.

    Functions and sources of event-related EEG alpha oscillations studied with the wavelet transform

    Clin. Neurophysiol.

    (1999)
  • V.J. Samar et al.

    Multiresolution analysis of event-related potentials by wavelet decomposition

    Brain Cogn.

    (1995)
  • S.J. Schiff et al.

    Fast wavelet transformation of EEG

    Electroencephalogr. Clin. Neurophysiol.

    (1994)
  • E.N. Sokolov

    Neuronal models and the orienting response

  • E. Basar

    EEG-Brain Dynamics. Relation between EEG and Brain Evoked Potentials

    (1980)
  • R.J. Barry et al.

    Does the N100 evoked potential really habituate? Evidence from a paradigm appropriate to a clinical setting

    Int. J. Psychophysiol.

    (1993)
  • Cited by (0)

    View full text