Auditory Systems Are Calibrated through Polymodal Integration

In Chapter 8 we describe the propensity of some brain regions to respond to inputs originating in two or more different sensory systems; we refer to this process as polymodal, or multisensory, integration. Experimental data indicate that information from other sensory modalities fine-tunes the accuracy of auditory systems.

Eric Knudsen et al. (1984) performed elegant studies using an especially acute binaural perceiver: the barn owl. When hunting, owls use both arrival differences and intensity differences to accurately localize sounds at night (Peña and Konishi, 2000). Cells of the avian tectum are arranged in a roughly spherical map of space (Knudsen, 1984; Knudsen and Konishi, 1978). In the owl tectum, both auditory and visual space are represented, and the maps for the two senses overlap very closely (Knudsen, 1982). In fact, most cells in the owl tectum are polymodal: they respond to both auditory and visual stimuli, and it is thought that this close alignment of auditory and visual maps of space helps guide behavioral responses toward stimuli.

To assess the impact of early experience on the development of these maps, investigators plugged one ear of adult and baby owls. These birds made large errors in localizing sounds, with responses shifted in the direction of the open ear—presumably because the sound was more intense in that ear, which would normally mean that it had come from that side. Adult animals never seemed to adjust to the earplug, but owls younger than 8 weeks at the time of plugging slowly began to compensate.

It turns out that vision is a key mediator in recalibrating auditory localization: the compensation did not occur if the owls were deprived of vision; and if they were fitted with prism glasses that deviated vision by 10° (Figure 1), the adjustment of auditory localization was matched to this visual error (Knudsen and Knudsen, 1985). Apparently the induced mismatch between the auditory and visual tectal maps provoked a remapping of auditory space in the baby owls that was no longer possible in the brains of the adult owls. These changes might arise either from structural modifications of growing neural circuits or from modulations of synaptic effectiveness (Knudsen, 1998). The tight integration of auditory and visual perception probably relates to higher-level mechanisms that direct attention to specific locations in the environment (Winkowski and Knudsen, 2006).

Figure 1  The Role of Vision in Auditory Localization
This young owl has had prisms fitted over its eyes to deviate vision by about 10° to the side. At first, the owl makes mistakes when reaching for a visual target, but eventually it learns to adapt to the prisms. The neurons in its tectum also adapt, so now neurons excited by a sound from a particular location in space are also excited by visual stimuli presented 10° to the left of that location. (Courtesy of Eric Knudsen.)

Although in owls only juveniles have auditory systems that can be remodeled through experience, the auditory areas of mammals and other birds show plasticity that persists into adulthood. So, for example, if adult monkeys are trained for several months to discriminate sounds in order to receive a food reward, the cortical representation for the training frequencies becomes substantially larger. Control subjects that just passively listen to the same tones do not show this response (Recanzone et al., 1993). Similarly, conditioning a guinea pig to tones of a particular frequency can cause cortical neurons to shift their response to favor that frequency (N. M. Weinberger, 1998). This remodeling can occur very quickly, on the order of just a few minutes (Fritz et al., 2003), reflecting an adaptive ability to continually tune and retune the auditory cortex to detect biologically significant sounds.

Intriguingly, reward may be a potent contributor to the remodeling of auditory cortex. Each day for 20 days, S. Bao et al. (2001) presented rats with tones paired with microstimulation of the ventral tegmental area (VTA). VTA stimulation activates the brain’s dopamine-based reward system (see Chapter 4). The investigators subsequently found that cortical representations were greatly increased for tones that occurred shortly before—but not after—VTA stimulation. From an evolutionary perspective, this sort of mechanism is highly adaptive because it specifically shapes the cortex to respond to stimuli that have previously signaled a reward, often in some other sensory modality.

Go