Evidence suggesting that infants appreciate a range of cross-sensory correspondences is growing rapidly (see Dolscheid, Hunnius, Casasanto & Majid, 2014; Fernández-Prieto, Navarra & Pons, 2015; Haryu & Kajikawa, 2012; Mondloch & Maurer, 2004; Walker, Bremner, Mason, Spring, Mattock, Slater, & Johnson, 2010; Walker, Bremner, Lunghi, Dolscheid, Barba & Simion, 2018), and yet there is no known attempt to establish the functional significance of these correspondences in infancy. Research shows that speakers manipulate their prosody (i.e. melody of spoken language) to communicate the meaning of unfamiliar words and do so in ways that exploit the cross-sensory correspondences between, for example, pitch and size (Nygaard, Herold & Namy, 2009) and pitch and height (Shintel, Nusbaum & Okrent, 2006). But do infants attend to a speaker’s prosody in this context to interpret the meaning of unfamiliar words? The aim of this thesis is to further establish how infant-directed speakers use prosody to communicate the cross-sensory meanings of words and, for the first time, identify whether infants capitalise on their sensitivity to cross-sensory correspondences to resolve linguistic uncertainty. In Experiment 1 – 4 we identify a list of novel pseudowords to use in all experiments being reported. These pseudowords were judged by participants as being neutral in terms of their sound-symbolic potential, allowing us to rule out the impact of sound-symbolism in our investigation. Experiment 5 provides support for earlier studies revealing cross-sensory correspondences in infant-directed speech. When presented with pseudowords spoken in a prosodically meaningful way, 13-month-old infants demonstrated a preference for objects that were contradictory to the cross-sensory acoustic properties of speech (e.g. lower-pitch voice with higher objects) (Experiment 6), and adults failed to match pseudowords with objects based on the prosodic information that was provided (Experiment 7). However, Experiment 8 provides evidence that 24-month-olds match pseudowords spoken in a higher-pitch voice, and at a faster rate, with objects that are visually higher in space. The implications of these findings are discussed, with suggestions as to how they can be usefully extended.