Abstract The correspondence between prosodic prominence and information-structural categories like focus has been investigated extensively for many years. The seminal work by Pierrehumbert & Hirschberg (1990), proposing a categorical distinction between H* and L+H* pitch accents, has inspired much subsequent work but the nature of this distinction is a matter of on-going debate. Some evidence that H* and L+H* accents have distinct interpretations has been put forward (e.g., Alter, et al., 2001; Krahmer & Swerts, 2001; Selkirk, 2002; Ito, et al. 2004). More recently, however, experimental studies have favored a gradient distinction between the two pitch accent types (e.g., Kügler & Gollrad, 2015; Grice et al., 2018). Concerning the corresponding interpretation, it has been proposed that H* and L+H* accents vary only in the salience or size of the alternative set (Watson, Gunlogson, & Tanenhaus, 2008; Calhoun, 2009) rather than representing different types of focus. The current talk will contribute this debate by investigating the role of focal pitch accents in the computation of implicature (inspired by Rooth, 1992 and subsequent work in formal semantics/pragmatics). First, I will present a series of experiments showing that the L+H* pitch accent makes a contrastive interpretation highly prevalent but that this interpretation is delayed in online language processing, compared to a condition that lexically encodes the same meaning (with the focus particle only). Furthermore, the results showed that when combined with the particle also, the L+H* facilitated the derivation of the additive meaning. In conclusion, the results suggest a dual function of focal pitch accents in (i) helping listeners to identify relevant alternatives in context and (ii) providing a strong cue that an inference should be derived. This role of prosody is context-dependent and interacts with lexical cues preceding the relevant pitch accents. Therefore, the results favor a view in which the L+H* pitch accent does not conventionally encode contrast and in which the distinction between L+H* and H* pitch accents is gradient. I will conclude by discussing how the role of focal pitch accents can be modeled in two competing formal accounts of implicature (Sauerland, 2004 and Chierchia, 2013). Overall, the findings suggest that focus intonation affects the computation of implicature beyond its role in generating alternatives.