Although children are competent at producing the adverbial connectives because and if from a young age (e.g. Diessel, 2004; Hood & Bloom, 1979), their ability to understand them in the speech of others is unreliable until much later in childhood (e.g. Emerson & Gekoski, 1980). However, corpus studies have also shown that there are pragmatic patterns in childrenâ€™s production of these connectives (De Ruiter, Lemen, Brandt, Theakston, & Lieven, in press), which are often overlooked in comprehension research. In particular, although Sweetser (1990) argued that these connectives can express three different pragmatic functions (Content, Epistemic, Speech-Act), previous research on childrenâ€™s understanding of these connectives has generally been limited to the Content type (e.g. De Ruiter, Theakston, Brandt, & Lieven, 2018; Emerson & Gekoski, 1980). While this aligns with accounts of cognitive complexity, which predict the Content type should be the easiest to acquire (e.g. Sanders, 2005; Zufferey, 2010), this seems to underestimate the influence of input patterns, which from a usage-based perspective (e.g. Tomasello, 2001), should impact language acquisition. For example, based on input frequency, the Speech-Act type should be the easiest type to acquire for because (e.g. De Ruiter et al., in press). Thus, to contribute to a better understanding of childrenâ€™s acquisition of these connectives, this thesis investigates the effects of this functional variation on childrenâ€™s production and comprehension of because and if, and also explores how acquisition of these pragmatic functions is related to complexity and input patterns. These aims are addressed via mixed measures (production, accuracy, response time, eye-tracking) over three separate studies. In Chapter 3, mothersâ€™ and childrenâ€™s because and if Speech-Act sentences are analysed for both form and function to provide more information about the types of Speech-Act sentences children produce and how these relate to input. In Chapter 4, childrenâ€™s comprehension of Content and Speech-Act because- and if-sentences are compared via accuracy and response time measures on a behavioural task. In analysing patterns associated with comprehension and processing of these two pragmatic types, evidence for both input and complexity are considered. In Chapter 5, eye-tracking, accuracy and response time data provide detailed information about childrenâ€™s ability to predict the functional meaning (Content, Epistemic and Speech-Act) signalled by these adverbial connectives. Overall, the data show that childrenâ€™s acquisition of the different pragmatic functions is highly influenced by patterns in the input, although the evidence for this varies with the methodology used. These findings are used to critically evaluate the cognitive complexity and usage-based approaches in terms of their role in explaining acquisiton of these connectives. The results are also discussed in terms of their implications on existing and future research in this area, as well as what they contribute to an understanding of childrenâ€™s pragmatic awareness.