Recently it has become increasingly apparent that primary sensory cortices, like the primary auditory field AI, are not purely unimodal but also process complex information from other sensory modalities as well as non-sensory information. Here, we will review our anatomical work on the connections of the auditory cortex in a small desert rodent, the Mongolian gerbil (Meriones unguiculatus), a common animal model in auditory research. In particular we will show that AI has multiple connections with auditory, non-auditory sensory (visual, somatosensory, olfactory), multisensory, motor, “higher order� associative and neuromodulatory brain structures. These connections are suitable to mediate multimodal integration processes as observed at the level of AI and, in turn, enable AI to influence other sensory and non-sensory systems at several cortical and subcortical levels.

The issue was approached by means of the axonal transport of various sensitive neuronal tracers, mainly by the bidirectional transport of fluorescein-labeled (FD) and tetramethylrhodamine-labeled dextran (TMRD), which were simultaneously injected into different frequency regions of the gerbil’s AI.

As expected, major connections of AI are with auditory structures like the other ipsilateral and contralateral auditory cortical fields and the auditory thalamus. However, approx. 18% of the inputs to AI arise from non-auditory cortical (e.g., posterior parietal cortex, perirhinal cortex) and subcortical areas (e.g., multisensory thalamic nuclei, neurotransmitter-related structures of the brainstem) as estimated by the number of retrogradely labeled cells in these brain regions.

The analysis of the topography of the FD- and TMRD-labeled cells of origin and axonal terminations, respectively, revealed that the connections between AI and the other auditory structures are usually tonotopically organized, whereas all other connections are non-tonotopic.

The laminar pattern of corticopetal, corticocortical and corticofugal connections suggests that AI receives primarily bottom-up-like inputs from the ascending auditory pathway and conveys auditory information feedforward (bottom-up) to its cortical target areas. In turn, AI receives largely cortical feedback inputs (top-down) and projects top-down-like to its subcortical targets. However, if one considers the organization of corticocortical and thalamo-cortico-thalamic feedback loops, this plain classification of the AI connectivities becomes increasingly complex.