The neural circuit for tone-specific plasticity in the auditory system elicited by conditioning

Auditory fear conditioning elicits plastic changes specific to a conditioning tonal stimulus (CS) in the central auditory system. Gao and Suga (1998) proposed the neural circuit for producing these tone-specific changes, represented by best frequency (BF) shifts. The Gao–Suga model (Fig. 1, solid arrows), elaborated upon by Suga et al. (2000, 2002), states that small (or subthreshold) short-term cortical and collicular BF shifts specific to tone bursts (CS) are evoked by the neural circuit within the auditory cortex (Chowdhury and Suga 2000; Sakai and Suga 2001, 2002; Xiao and Suga 2002; Yan and Ehret 2002; Ma and Suga 2003, 2004) and corticofugal feedback loops (Gao and Suga 1998, 2000) activated by CS alone, and this cortical BF shift is augmented and changed into the long-term BF shift by acetylcholine released into the auditory cortex from the cholinergic basal forebrain, the nucleus basalis (Bakin and Weinberger 1996; Bjordahl et al. 1998; Kilgard and Merzenich 1998; Ma and Suga 2003; Yan and Zhang 2005; Zhang et al. 2005). In this model, the nucleus basalis is activated by the auditory and somatosensory cortices (Gao and Suga 1998, 2000; Ma and Suga 2001, 2003; Ji and Suga 2007b) via the association cortex and amygdala where CS is associated with an unconditioned leg stimulus (US). In addition, CS–US association may also occur in the association cortex. The collicular BF shift is increased by the augmented cortical BF shift through corticofugal feedback and contributes to the development of the large long-term cortical BF shift (Ji et al. 2001; Ma and Suga 2005). The ascending and descending (corticofugal) systems form positive feedback loops for the BF shifts. The gain of these feedback loops is presumably controlled by the thalamic reticular nucleus. (For simplicity, the corticothalamic projection is not described in the Gao–Suga model.) This model is fundamentally different from the Weinberger model (Weinberger 1998, 2007). It proposes that the small, short-lasting cortical BF shift is evoked by the neural net intrinsic to the central auditory system without CS–US association, whereas the Weinberger model proposes that it is evoked by the multisensory thalamic nuclei (the medial division of the medial geniculate body [MGBm] and the posterior interlaminar nucleus [PIN]) only after CS–US association occurs in these nuclei. Many neurophysiological findings indicate that the cortical and collicular BF shifts are evoked without CS–US association in the MGBm and PIN (Bakin and Weinberger 1996; Bjordahl et al. 1998; Kilgard and Merzenich 1998; Chowdhury and Suga 2000; Ma and Suga 2001, 2003, 2005; Sakai and Suga 2001, 2002; Xiao and Suga 2002; Yan and Ehret 2002; Yan and Zhang 2005; Zhang and Suga 2005; Zhang et al. 2005; Wu and Yan 2007) and that the inferior colliculus shows conditioningelicited BF shifts (Gao and Suga 1998, 2000; Ji et al. 2001, 2005; Ji and Suga 2003, 2007b). The Gao–Suga model is based on recent neurophysiological findings (Suga and Ma 2003), whereas the Weinberger model (Weinberger 2007, Fig. 12) does not incorporate any of these findings. Therefore, it is incorrect or incomplete. Suga does not doubt that MGBm neurons show responses to both CS (tone) and US (foot shock), BF shifts, and long-term potentiation or that they project to layer I in the auditory cortex (Weinberger 2007). However, the MGBm activation and inactivation experiments have not yet been performed to prove that the MGBm evokes the cortical BF shift. The serious problem is Weinberger’s interpretation of the role of the MGBm in evoking the cortical BF shift. Suga (2008) hypothesizes that the MGBm is involved in the nonspecific plasticity (nonspecific augmentation) of cortical neurons elicited by unpaired CS and US (Fig. 1, dashed arrows), not in the tone-specific plasticity (BF shift) elicited by paired CS and US. A conditioned big brown bat shows body movements (Gao and Suga 1998) and a heart-rate decrease (Fig. 2; Ji and Suga 2007a) to the conditioned tonal stimulation (CS). The noticeable cortical and collicular BF shifts are evoked neither by short acoustic stimulation (CS) alone nor by electric leg stimulation (US) alone but by paired stimulation (CS–US). Backward conditioning (US–CS) does not evoke these BF shifts. The BF shifts are specific to the frequency of the CS. Unlike the collicular BF shift, the cortical BF shift gradually increases and reaches a plateau after the termination of the conditioning. This plateau is sustained for a long time (Gao and Suga 2000; Suga et al. 2000; Ji et al. 2001). This long-term cortical BF shift is elicited by an increase in the cortical acetylcholine level resulting from activation of the nucleus basalis (Ji et al. 2001; Ji and Suga 2003; Ma and Suga 2003). Therefore, all these findings of the cortical BF shift in the big brown bat are the same as the findings in rodents and meet “all the criteria for physiological plasticity induced by the associative process” proposed by Weinberger (2004).