Adherence to a strict gluten-free diet is the only treatment for coeliac disease. Nonetheless, many individuals with the disease struggle to achieve and maintain strict adherence. While the theory of planned behaviour is useful for predicting gluten-free diet adherence, an intention-behaviour gap remains. The aim of this study was to investigate the roles of habit and perceived behavioural control in moderating the intention-behaviour relationship in gluten-free diet adherence. A significant three-way interaction was found such that the association between intention and adherence was dependent on both perceived behavioural control and habit. Implications for both theory and intervention design are discussed.

a b s t r a c tThe aims of this paper were to systematically review the evidence for the effectiveness of active behaviour change safety interventions in the construction industry; and to determine the intervention characteristics most commonly associated with effectiveness in reducing injury rates and improving safety behaviour -intensity/frequency/duration, behaviour change techniques (BCTs) and theory-base. An electronic literature search (June 2014) was conducted to identify eligible interventions: those involving active involvement from workers/management in the construction industry; targeted one/both of the primary outcomes. All intervention designs involving construction workers aged >18 years, published in English and in a peer-reviewed journal were included. Fifteen studies were included, half of which successfully improved injury rates. Longer interventions and those that included active/volitional BCTs (feedback/monitoring rather than instruction/information) were more effective. The methodological quality of the interventions was poor and use of theory was inconsistent and infrequent. Despite some positive results, very few interventions achieved all their aims. More rigorous, theory-driven research is needed to structure intervention efforts and determine the mechanism of action of effective interventions.

Objective: Alcohol-involved sexual assault remains a pervasive problem, with extensive individual-and societal-level costs. Emotion regulation (ER), the process through which an individual modulates emotional states, remains an understudied predictor of sexual assault perpetration, with the past research focusing on general ER tendencies (e.g., trait ER) as predictors of sexual assault perpetration. This study sought to examine the associations between state ER on sexual assault perpetration in the context of state anger and acute alcohol intoxication. Method: Single, male social drinkers aged 21-30 with a history of sexual risktaking (N = 92) participated in an alcohol administration paradigm and were randomly assigned to an alcohol condition [sober control vs. intoxicated (BrAC = .10%)]. Intoxicated and sober participants completed a sexual assault analogue that assessed state anger, state ER, and sexual assault perpetration intentions against a hypothetical female partner. Results: Path analysis demonstrated interactive effects of state ER and state anger on sexual assault perpetration intentions. Relative to men with low and moderate levels of anger, state ER was associated with lower intentions to perpetrate sexual assault for men with high levels of anger. Alcohol intoxication did not directly predict state ER, state anger, or sexual assault perpetration intentions. Conclusions: The results suggested that state ER may be protective against sexual assault perpetration for men who experience anger in response to a partner's expression of nonconsent. Because replication is necessary, the results carry tentative implications for state ER as an intervention target for sexual assault prevention programming.

A test was made of the contrary predictions of Spence and Estes in the type of situation in which two competing responses are reinforced in the ratio of 2: I through forced trials and measures of performance are taken on free trials where S may make either response. Whereas the two responses differed most in strength at an early stage in the training of a group of 40 human Ss, the performance curve for free trials did not rise to a maximum at this stage and then fall, as S ence has predicted, but simply ap eared to approach the 67 per cent leve P to be expected in terms of Estes' P ormulation.The theories of learning proposed by Spence (1956) and Estes (1959) appear to generate contrary predictions under certain conditions in the type of situation in which (u) two competing responses are reinforced with fixed differential frequencies through the inclusion of forced triaIs, and (b) measures of performance are taken on free trials where

No abstract