Having briefly reviewed implicit measures, including the Implicit Relational Assessment Procedure (IRAP), and examined the small number of published studies that have explored the possible role of methodological features of the task itself, the current thesis set about investigating other procedural parameters that may influence IRAP effects. The increasing use of the IRAP in clinically-relevant domains made this task seem even more pressing and necessary. At one level, we simply hoped to contribute to the small literature on the possible influence of these factors on IRAP outcomes. At another level, we hoped that these exploratory methodological analyses would help to pave the way for more robust use of the IRAP in applied and clinical domains. The current thesis comprised of four studies (Experiments, 1, 2, 3a and 3b). In order to permit useful comparisons of the response patterns of the various procedural manipulations, there are strong overlaps in the experimental designs and analytic strategies employed across the four studies. Namely, all four studies involve a Typical IRAP as a control procedure and this is systematically compared to IRAPs with specific procedural modifications. In Experiment 1 (Chapter 2), we compared a Typical IRAP with a Natural Language IRAP in which the label and target stimuli were combined, rather than presented separately as is usually the case. This manipulation was simply to determine whether the use of more complex label-target combinations (that would perhaps be better suited to clinical research) would produce different IRAP effects than those typically observed with separate label and target stimuli. Overall, patterns of responding were very similar across both IRAPs and the findings, therefore, from Experiment 1 supported existing evidence which suggests that combining the label and target stimuli produces sound IRAP effects. However, such a manipulation involving the labels and targets appeared to have little or no direct impact on the outcomes. In Experiment 2 (Chapter 3), we employed a Self-esteem IRAP and manipulated the order in which the self vs. others rules were presented. In the Self-Rule First Condition, participants were presented with the rules as follows: the self-positive rule followed by the others-negative rule, while in the Others-Rule First Condition participants were presented with the others-positive rule followed by the self-negative rule. This manipulation sought to determine whether the simple order in which the rules are presented influenced IRAP outcomes. The results indicated that patterns of responding appeared to differ somewhat, but not significantly, between conditions, especially on Self-Negative and Others-Positive trial-types. Overall, the findings from Experiment 2 supported some existing evidence which suggests that rules exert some influence on IRAP effects, and thus warrant empirical attention. In Experiment 3a (Chapter 4), we employed the Spider Fear IRAP and manipulated the length of the test blocks presented between rules. To do so, we compared a Fast Switching IRAP and a Typical Switching IRAP. Specifically, the Fast Switching IRAP halved the length of the test blocks between rules, relative to the Typical Switching IRAP (and IRAPs generally), such that each block now contained only 16 trials, and not 32. Again, this manipulation sought to determine whether the typical rate at which blocks switch exerts its own influence on outcomes normally observed with the IRAP. Patterns of responding were somewhat similar across the two IRAPs. Stronger responding was recorded on the Fast Switching IRAP on the Fear-Spiders and Approach-Spiders trial-types, while the Typical Switching IRAP was stronger on the Fear-Nature and Approach-Nature trial-types. The difference on Fear-Spiders was significant. While there were differences in the two outcomes, one could not argue that either condition yielded a better IRAP outcome than the other. In Experiment 3b (Chapter 4), we again employed the Spider Fear IRAP and now manipulated the number of the test blocks presented. To do so, we compared a Double Length IRAP and a Typical Length IRAP. That is, instead of the typical 6 test blocks (i.e. 3 per rule), the Double Length IRAP presented 12 test blocks (6 per rule), although each test block contained the typical 32 trials. Overall, therefore, this modified IRAP presented a total of 384 trials -- that is double the number of trials presented in the Typical Length IRAP (i.e. 192). Again, this manipulation sought to determine whether the typical number of blocks exerts its own influence on outcomes normally observed with the IRAP. Patterns of responding were very similar across the Typical and Double Length IRAPs. In both cases, participants showed consistent responding with increasing numbers of block pairs, with no significant differences across each trial-type of increasing block pairs. The current thesis, therefore, undertook a number of systematic manipulations of various procedural parameters of the IRAP. Overall, one must conclude that these exerted practically no significant influence on the observed IRAP effects, although some variations were detected. These changes are discussed in the context of the existing literature and particularly with regard to more extensive use of the IRAP in the future.