Interest in iconicity (the resemblance-based mapping between aspects of form and meaning) is in the midst of a resurgence, and a prominent focus in the field has been the possible role of iconicity in language learning. Here we critically review theory and empirical findings in this domain. We distinguish local learning enhancement (where the iconicity of certain lexical items influences the learning of those items) and general learning enhancement (where the iconicity of certain lexical items influences the later learning of non-iconic items or systems). We find that evidence for local learning enhancement is quite strong, though not as clear cut as it is often described and based on a limited sample of languages. Despite common claims about broader facilitatory effects of iconicity on learning, we find that current evidence for general learning enhancement is lacking. We suggest a number of productive avenues for future research and specify what types of evidence would be required to show a role for iconicity in general learning enhancement. We also review evidence for functions of iconicity beyond word learning: iconicity enhances comprehension by providing complementary representations, supports communication about sensory imagery, and expresses affective meanings. Even if learning benefits may be modest or cross-linguistically varied, on balance, iconicity emerges as a vital aspect of language.

Comparative perspectives on primate and human communication have been marked by two equally untenable extremes: either language is special, without significant evolutionary precedent, or it is not: it is continuous in most aspects with animal communication systems. In this article we outline fertile common ground and point towards synthetic approaches that can unify the study of human and animal communication. First, we suggest that humans have a large suite of perceptual biases that introduce a pressure for languages to be 'functionally deployable'. We suggest that human languages are shaped by this pressure, along with previously established pressures to be both learnable and compressible, and domain-general constraints like memory. Collectively, we suggest that non-arbitrary structure-function relationships are crucial for the deployment of language and communication systems more generally.

Comparative research on communication systems in human and non-human primates has struggled to find a common language for comparisons. Research has too often either embraced application of linguistic constructs to animal signals, or eschewed them altogether, with the result being either uncritical acceptance of continuity between human and animal communication, or wholesale dismissal of it. This chapter attempts to better align the discussion of communication in the two groups by highlighting the importance of low-level features of communication indisputably common to both. From this common starting point, the authors of this chapter invoke functional deployability as a critical pressure influencing signal structure and use, emphasizing conserved perceptuocognitive mechanisms as common targets of signal action. In recognizing these foundational building blocks, many productive avenues for future research open up. In particular, the authors suggest that perceptual and cognitive biases that promote certain types of signal-meaning mappings provide a natural, biosemantic substrate on which more robust communicative systems can be built.