Learning and Motivation 63 (2018) 37–48

Contents lists available at ScienceDirect

Learning and Motivation

journal homepage: www.elsevier.com/locate/l&m

Working for beverages without being thirsty: Human Pavlovian- instrumental transfer despite outcome devaluation T ⁎ Matteo De Tommasoa, , Tommaso Mastropasquaa, Massimo Turattoa,b a Center for Mind/Brain Sciences, University of Trento, Corso Bettini, 31, 38068, Rovereto, Italy b Department of Cognitive Psychology, University of Trento, Corso Bettini, 31, 38068, Rovereto, Italy

ARTICLE INFO ABSTRACT

Keywords: The incentive-motivational salience acquired by a Pavlovian conditioned stimulus (CS) is re- PIT flected by its ability to strengthen the performance of a separately learned instrumental action Incentive salience exerted to obtain an outcome, a phenomenon known as Pavlovian-Instrumental transfer (PIT). By Wanting means of a PIT paradigm, the present study addressed whether the CS motivational properties Reward devaluation vary dynamically with the value of the associated outcome. Previous studies on human PIT and Physiological state outcome devaluation have provided mixed results, showing that in some cases post-training devaluation leaves PIT unaffected when outcomes are palatable foods or drugs, and when the devalued outcome is not consumed immediately. In Experiment 1, thirsty participants first learned to squeeze a rubber bulb to accumulate a beverage (plain water or sugary drink); then participants learned Pavlovian associations between cues and the beverage. When tested in ex- tinction, a PIT effect emerged as expected. In Experiment 2, the PIT effect emerged even despite participants quenched their thirst before the test phase. Our results suggest that the incentive properties of a CS can surprisingly and irrationally endure the devaluation of the associated outcome even when plain water is used as reward, and thirst is quenched by immediate reward consumption. This result may provide important insights in the understanding of the psycholo- gical mechanisms underlying different types of addiction.

1. Introduction

Conditioning and other learning mechanisms often play a fundamental role in determining the consumption of food and other substances, but are also involved in maladaptive motivational behaviors such as compulsive food seeking and addiction. A key notion of the incentive sensitization theory (Robinson & Berridge, 2008; Robinson, Robinson, & Berridge, 2013;), one of the most recent theories of addiction, is that the conditioned stimulus (CS) can acquire the motivational properties of the unconditioned stimulus (US) through Pavlovian learning. This view stems from the incentive theories of motivation originally proposed by Bindra (Bindra, 1978) and Toates (Berridge, 2001; Toates, 1986), and postulates that in addition to be a predictor of the US occurrence, the CS can also trigger the desire for the associated US (Berridge, 2012). Furthermore, because of such features the CS would become a motivational magnet, thus attracting attention and inciting action. The motivational properties of the CS are revealed by the Pavlovian-Instrumental transfer (PIT) effect, a phenomenon showing how reward cues can influence behavior. The PIT rests on the interaction between Pavlovian and Instrumental conditioning, and shows the influence of a Pavlovian CS on the instrumental action. Hence, the PIT effect can reveal how a CS-triggered motivation may

⁎ Corresponding author. E-mail addresses: [email protected] (M. De Tommaso), [email protected] (T. Mastropasqua), [email protected] (M. Turatto). https://doi.org/10.1016/j.lmot.2018.01.001 Received 16 October 2017; Received in revised form 13 January 2018; Accepted 16 January 2018 0023-9690/ © 2018 Elsevier Inc. All rights reserved. M. De Tommaso et al. Learning and Motivation 63 (2018) 37–48 affect the response provided to obtain the reward. Although the PIT has been traditionally and extensively studied in animals, it has recently been reported also in humans (for a review and a meta-analysis see, Cartoni, Balleine, & Baldassarre, 2016; Holmes, Marchand, & Coutureau, 2010). According to the associative–cybernetic model of the PIT effect, the CS would activate the outcome representation, which in turn promotes the instrumental action through an association learned during the instrumental conditioning phase (Balleine & Ostlund, 2007). The Konorskian view of PIT, assumes that different Pavlovian CSs, which are associated with different rewards, are able to activate a common motivational system (Dickinson & Balleine, 2002; Konorski, 1967). The PIT has thus been interpreted as a measure of incentive salience (Berridge, 2001; Berridge & Robinson, 2003; Pool, Brosch, Delplanque, & Sander, 2015; Pool, Sennwald, Delplanque, Brosch, & Sander, 2016). An interesting and still open theoretical question emerging from the incentive sensitization theory is whether the learned in- centive properties of the CS change as a function of the motivational value of the associated US (Cartoni et al., 2016). A possible answer comes from the incentive salience model outlined by Zhang and colleagues (Zhang, Berridge, Tindell, Smith, & Aldridge, 2009), which postulates that the magnitude of the CS incentive salience or “wanting” is determined by two main factors: the CS associative history with the US, and the physiological state (K) of the organism relative to the US when the CS is encountered. These factors act in concert in determining the CS incentive salience, so that certain levels of hunger, drug states or stress (K) are able to transform the CS in a powerful motivational trigger. Given that the CS incentive properties vary with K level, the model predicts that an upward shift of K would translate into an increased CS-triggered “wanting”, while a downward shift of K would cause a decrease of the CS incentive salience. This prediction, however, is only partly confirmed by previous PIT studies that have modulated the US value. In agreement with an upshift modulation predicted by the incentive salience model, when human participants were exposed to a stressful experience, the PIT effect caused by a CS for a chocolate odor was amplified (Pool et al., 2015). Instead, evidence of a downshift modulation emerged in a study in which human participants engaged in a “stock market game” did not increase the effort they made to obtain a depreciated currency when shown a previously learned CS for the same currency, thus reporting a successful outcome devaluation effect (Allman, DeLeon, Cataldo, Holland, & Johnson, 2010; Eder & Dignath, 2016a). However, such deva- luation procedures might have been effective on the PIT because of the specific paradigm used. In other words, a currency deva- luation procedure in a “stock market game” could be interpreted as a cognitive rule as compared to a devaluation emerging from a motivational satiation procedure based on primary rewards (Cartoni et al., 2016). Another element to consider when addressing the differential effects of the US devaluation on PIT is suggested by the study of Eder and Dignath (2016b). When the devaluation of a primary reward was achieved by a taste aversion procedure, its consequence on PIT was effective only if participants consumed the US immediately after earning it. However, a different scenario is suggested by a series of humans and animals studies reporting that a CS retains its motivational control over instrumental responses (i.e., the PIT effect is unaffected) even if the corresponding US is devalued (Colagiuri & Lovibond, 2015; Colwill & Rescorla, 1990; Corbit, Janak, & Balleine, 2007; Hogarth & Chase, 2011; Holland, 2004; Watson, Wiers, Hommel, & De Wit, 2014). In particular, the presence of the PIT effect after reward devaluation has been interpreted as a persistent CS-triggered motivational response (Berridge, 2012). This may be partially explained by the fact that some studies have used rewards whose incentive value cannot be easily diminished by means of satiation or single-taste aversion procedures. So, for example, a highly palatable food (e.g., chocolate or candies) as well as drugs (e.g., tobacco) could retain, at some level, the ability to activate moti- vational states irrespective of the organism’s satiation level (Hogarth, 2012; Hogarth & Chase, 2011). To put it differently, although a piece of cake is extremely motivationally salient when we are hungry, the same cake can still be desirable even after a full meal. A recent study supports this possibility by showing that rats fail to exhibit devaluation effects in contexts paired with junk food as compared to contexts paired with less palatable regular chow (Kendig, Cheung, Raymond, & Corbit, 2016). In addition, although satiation was shown to specifically reduce the desire of an outcome in some studies, satiation might not have been always drive- specific. So, for example, a previous study (Watson et al., 2014) showed that cues anticipating chocolate and popcorn increased the subsequent instrumental responding for the specific previously paired food (i.e., PIT), but satiation of one of the two foods failed to reduce the PIT effect. The apparent failure of reward devaluation can be explained by assuming that the food cue was effective because the food-specific satiation did not reduce the general hunger drive enough. In light of this intricate scenario, our aim was to clarify how changes in the US value may affect the corresponding CS incentive salience. With this goal in mind we tested human participants in a PIT paradigm defined by two key features: first, we used a primary reward; second, the reward used could be easily devalued, thus changing the corresponding incentive motivation. In Experiment 1, thirsty participants first learned to squeeze a rubber bulb to gain a liquid reward. Then, they were administered a Pavlovian con- ditioning in which three CSs were paired with different probabilities of liquid delivery. Finally, participants were tested for PIT. They performed the instrumental action learned in the first phase in the presence of one of the three possible CSs. The prediction was that the strength of the PIT effect would have been proportional to the CS-US contingency: p = 0.2, p = 0.5 and p = 0.8 (Trick, Hogarth, & Duka, 2011). In other words, the performance triggered by the CS should be greater for a CS that was a highly reliable predictor of the US (i.e., p = 0.8) as compared to a CS that predicted the US with a low probability (i.e., p = 0.2). Note that in a standard PIT paradigm, the PIT effect is attested by an increased rate of instrumental responding in the presence of the CS previously associated with the reward (e.g., p = 1 or 0.8) as compared to a baseline condition in which either no CS or a CS never paired with the reward is presented (p = 0). Since in the present study we used three CSs with three different levels of contingency, a differential instrumental responding triggered by the CSs in the PIT phase must be taken as evidence of the PIT effect. In other words, the rate of instrumental responding with the CS associated with the lowest contingency (p = 0.2) served as baseline for the PIT effect. Importantly, participants were free to choose their preferred liquid US between plain water and well-known sugary beverages. Plain water is not necessarily an intrinsic desirable stimulus, because its desirability is proportional to the organism’s level of thirst. By contrast, sugary beverages could be more resistant to devaluation by thirst quench because of the addictive-like properties of sugar

38 Download English Version: https://daneshyari.com/en/article/7275675

Download Persian Version:

https://daneshyari.com/article/7275675

Daneshyari.com