A reappraisal of successive negative contrast in t

Authors: Stefanie Riemer Sarah L H Ellis Sian Ryan Hannah Thompson Oliver H P Burman

Publish Date: 2016/01/07

Volume: 19, Issue: 3, Pages: 471-481

PDF Link

Abstract

When an anticipated food reward is unexpectedly reduced in quality or quantity many mammals show a successive negative contrast SNC effect ie a reduction in instrumental or consummatory responses below the level shown by control animals that have only ever received the lowervalue reward SNC effects are believed to reflect an aversive emotional state caused by the discrepancy between the expected and the actual reward Furthermore how animals respond to such discrepancy has been suggested to be a sign of animals’ background mood state However the occurrence and interpretation of SNC effects are not unequivocal and there is a relative lack of studies conducted outside of laboratory conditions Here we tested two populations of domestic dogs 24 owned pet dogs and 21 dogs from rescue kennels in a SNC paradigm following the methodology by Bentosela et al J Comp Psychol 123125–130 2009 using a design that allowed a within as well as a between subjects analysis We found no evidence of a SNC effect in either population using a within or betweensubjects design Indeed the withinsubjects analysis revealed a reverse SNC effect with subjects in the shifted condition showing a significantly higher level of response even after they received an unexpected reduction in reward quality Using a within rather than a between subjects design may be beneficial in studies of SNC due to higher sensitivity and statistical power however order effects on subject performance need to be considered These results suggest that this particular SNC paradigm may not be sufficiently robust to replicate easily in a range of environmental contexts and populationsMany mammals will show a reduction in instrumental or consummatory responses when they experience an unexpected shift from a higher to a lower quality and/or quantity of reward relative to a control group that is exposed only to the lowerlevel reward Papini et al 1988 Flaherty 1999 Mustaca et al 2000 Bergvall et al 2007 Catanese et al 2011 If the responses of the ‘downshifted’ subjects fall below those of animals who have only ever received the less preferred reward ‘unshifted’ subjects the phenomenon is known as a successive negative contrast SNC effect see Flaherty 1999 for review SNC effects have been found in some mammalian species including in rats Rattus norvegicus eg Crespi 1942 Mellgren 1972 Pellegrini and Mustaca 2000 mice Mus musculus Mustaca et al 2000 sheep Ovis aries Catanese et al 2011 Greiveldinger et al 2011 fallow deer Dama dama Bergvall et al 2007 two marsupials Lutreolina crassicaudata and Didelphis albiventris Papini et al 1988 and domestic dogs Canis familiaris Bentosela et al 2009 With the exception of the starling Sturnus vulgaris Freidin et al 2009 to date investigation of SNC in other vertebrate species has given negative results for example in pigeons Columba livia Papini 1997 toads Bufo arenarum Muzio et al 1992 Papini et al 1995 turtles Geoclemys reevesii Papini and Ishida 1994 and goldfish Carassius auratus Lowes and Bitterman 1967 Couvillon and Bitterman 1985 However the effect has been observed in honey bees Bitterman 1976 Wiegmann and Smith 2009 and bumble bees Waldron et al 2005The exaggerated change in behavioural response eg reduced operant behaviour or reward consumption as a result of an unexpected reduction in reward value implies that animals form reward expectations and compare the quantity or quality of the present reward with those received previously Flaherty 1999 and it is suggested that they experience shortterm aversive emotions ie brief transient and stimulusdependent affective states eg ‘disappointment’ if these expectations are not met reviewed in Rosas et al 2007 Justel et al 2014 Papini 2014 How animals respond to such an unexpected reward reduction in terms of either the strength or duration of their response may also be dependent on their longerterm mood ie enduring and stimulusindependent affective states such as ‘depression’ Consequently it has been suggested that SNC might be valuable as a way of informing us about the background affective state of subjects as well as directly inducing affect Burman et al 2008 Mitchell et al 2012 Accordingly rats of strains selected for high emotional reactivity show enhanced SNC effects eg Cuenya et al 2012 Freet et al 2006 Gómez et al 2009 Ortega et al 2014 Rosas et al 2007Burman et al 2008 Mitchell et al 2012 and Chaby et al 2013 tested the effect of environmental manipulations on SNC in laboratory rats In Burman et al 2008 rats from unenriched housing assumed to be experiencing a more negative affective state displayed a prolonged response expressed as slower running speeds in a runway to an unexpected decrease in reward quantity compared to enriched rats indicating enhanced sensitivity to reward loss Burman et al 2008 Along similar lines rats that had experienced unpredictable and stressful environments during adolescence responded more strongly to a reward downshift than a control group in Chaby et al 2013 However Mitchell et al 2012 found that rats kept in barren environments showed an attenuated SNC effect compared to individuals from enriched housing Mitchell et al 2012 The authors suggested that the apparent contradiction in results compared to the study by Burman et al 2008 and Chaby et al 2013 could be explained by the possibility that access to the test chamber itself and the contrast to their unenriched home environment induced a positive affective state in the unenriched subjects since they had experienced daily rewardbased training in this location This reflects findings in other studies using cognitive approaches to assess affective state in animals in which the affective state at the moment of testing may differ from that predicted a priori eg Doyle et al 2010 Burman et al 2011 Thus the occurrence and interpretation of SNC effects are not unequivocal and more work is needed to validate its robustness as a measure of affective state in animals across different environmentsWhile SNC has been extensively investigated in the laboratory rat less work has investigated responses to reward downshifts in other mammals reviewed above With few exceptions the tested individuals came from highly standardised laboratory conditions and it is thus questionable whether results can be generalised to more heterogeneous environments outside of the laboratoryFor a number of reasons the domestic dog is a particularly valuable model species in which to advance the study of SNC as a measure of affective state Firstly dogs are a remarkably varied species residing in an extensive range of environments Taylor and Mills 2007 For example purposebred laboratory dogs represent a homogenous population residing in a homogeneous environment whereas owned pet dogs represent a heterogeneous population residing in a heterogeneous environment Secondly numerous studies have demonstrated advanced cognitive abilities in the domestic dog reviewed in Bensky et al 2013 thereby making it a useful species to investigate SNC where experimental paradigms often involve elements of training Finally many dogs find themselves living temporarily or permanently in kennels an environment which can result in poor welfare and negative affect Hennessy et al 1998 Coppola et al 2006 Taylor and Mills 2007 due to a number of factors such as physical confinement Wells 2004 and limited intra and interspecific contact for example isolation from a former attachment figure Tuber et al 1999 Such an environment therefore provides an opportunity to extend previous findings of the influence of affective manipulations on SNC eg Burman et al 2008 to outside the laboratory thereby further validating its reliability as a measure of affective state and ultimately animal welfareA pioneering study on SNC in domestic dogs conducted by Bentosela et al 2009 involved training pet owned dogs to receive a food reward in return for directing their gaze towards the experimenter’s eyes One group of dogs were shifted from the high value food reward dried liver to the low value reward dog food pellets while the other group always received the low value food Duration of eye gaze and proportion of food rejections in the two shift groups were compared While the effect of shift condition on duration of eye gaze was not significant Bentosela et al’s 2009 results did suggest the occurrence of a SNC effect as manifested in the observed rate of food rejection Conversely Pongrácz et al 2013 found no evidence of an incentive contrast effect when pet dogs were switched from sausage as a reward to carrot in a pointing taskThus previous findings regarding the occurrence of a SNC effect in domestic dogs are inconclusive Dogs’ strong reliance on human actions Topál et al 1997 Udell and Wynne 2008 or their interpretation of the pointing gesture as a command eg PratoPrevide et al 2008 but see Scheider et al 2013 may account for the lack of a SNC effect in the study by Pongrácz et al 2013 Given the small sample size in Bentosela et al 2009 13 subjects 7 in the shifted group and 6 in the unshifted group it is possible that clearer effects would emerge with a larger sample size due to higher statistical power It is also conceivable that random variation in their betweensubjects design may have reduced the effect given the large interindividual variability in both duration of eye gaze and food rejection rates This could be addressed by conducting the experiment using a withinsubjects design so that each subject acts as its own control cf Keren and Lewis 2014

Keywords: