Contrast effects in backward evaluative conditioning: Exploring effects of affective relief/disappointment versus instructional information.

Abstract

Past studies of backward evaluative conditioning (EC) have found an assimilation effect, in that neutral conditional stimuli (CS) were found to acquire the valence of co-occurring unconditional stimuli (US). Recent studies employing a concurrent forward and backward conditioning paradigm with instructions suggesting a contrastive relation between the US and the backward CS have resulted in contrast effects, in that backward CSs acquired valence opposite to the US. The current research investigated whether these effects were in fact due to the instructions highlighting the contrastive relation between the US and CS, or whether affective relief/disappointment experienced at US offset could account for this result. Consistent with the hypothesized role of instructions, backward CS contrast effects occurred only when instructions highlighted the valence of the US and attributed control of that US to the CSs. In contrast to the affective relief/disappointment hypothesis, no backward CS contrast effects were found without such instructions.

Startle during backward evaluative conditioning is not modulated by instructions

“…Explicit valence ratings revealed the same pattern of backward CS contrast effects in the start-stop instructions group as previously reported in studies using similar instructions, while no such finding emerged in the observe instructions group (Green et al, 2019;Moran & Bar-Anan, 2013;Moran et al, 2016). These findings again confirm that instructions emphasizing CS agency are required for the emergence of backward CS contrast effects on explicit valence ratings (unless shock USs are employed as in Andreatta et al, 2013).…”

“…The acquisition of valence can be measured by means of self‐report, where participants are asked to rate how pleasant or unpleasant they find a given stimulus (i.e., explicit valence ratings). Explicit valence ratings following forward (CS‐US) and backward (US‐CS) conditioning without instructions about CS/US relations reveal that the CS acquires valence in the direction of the US (an assimilation effect; Green, Luck, Gawronski, & Lipp, 2019; Mallan et al., 2008; see Hofmann et al., 2010 for a review and meta‐analysis). When instructions highlight that each set of CSs either starts or stops a pleasant or unpleasant US, backward CSs presented with positive USs are rated as less pleasant than backward CSs presented with negative USs (a contrast effect; Moran & Bar‐Anan, 2013; Green et al, 2019).…”

“…Explicit valence ratings following forward (CS‐US) and backward (US‐CS) conditioning without instructions about CS/US relations reveal that the CS acquires valence in the direction of the US (an assimilation effect; Green, Luck, Gawronski, & Lipp, 2019; Mallan et al., 2008; see Hofmann et al., 2010 for a review and meta‐analysis). When instructions highlight that each set of CSs either starts or stops a pleasant or unpleasant US, backward CSs presented with positive USs are rated as less pleasant than backward CSs presented with negative USs (a contrast effect; Moran & Bar‐Anan, 2013; Green et al, 2019). Moran and Bar‐Anan (2013) demonstrated this in a concurrent forward and backward evaluative conditioning paradigm (CS‐US‐CS) employing pleasant and unpleasant sound USs, and Green et al (2019) confirmed that the instructions highlighting the roles of the CSs were required for backward CS contrast effects to emerge in a picture‐picture paradigm.…”

“…When instructions highlight that each set of CSs either starts or stops a pleasant or unpleasant US, backward CSs presented with positive USs are rated as less pleasant than backward CSs presented with negative USs (a contrast effect; Moran & Bar‐Anan, 2013; Green et al, 2019). Moran and Bar‐Anan (2013) demonstrated this in a concurrent forward and backward evaluative conditioning paradigm (CS‐US‐CS) employing pleasant and unpleasant sound USs, and Green et al (2019) confirmed that the instructions highlighting the roles of the CSs were required for backward CS contrast effects to emerge in a picture‐picture paradigm. Contrast effects have also been demonstrated following forward conditioning when information that changes the relationship between the CS and US is presented.…”

“…Moreover, Green et al (2020a) used sound USs, which are arguably less intense than the electro-tactile USs used by Andreatta et al (2013). The fact that, in absence of explicit instructions, Green et al (2019) found no evidence of relief and disappointment learning on explicit ratings using picture USs while Andreatta et al (2013) did when using an electro-tactile US, suggests that a certain threshold of US intensity may be required for relief and disappointment learning to occur. Thus, it is currently unknown whether the backward CS contrast effect reported by Green et al (2020a) for explicit ratings and startle modulation is a true reflection of relief and disappointment learning, or the result of the instructional manipulation highlighting that the role of each set of CSs is to either start or stop the US.…”