Evidence that instrumental conditioning requires conscious awareness in humans.

Instrumental conditioning is a crucial substrate of adaptive behaviour, allowing individuals to selectively interact with the stimuli in their environment to maximise benefit and minimise harm. The extent to which complex forms of learning, such as instrumental conditioning, are possible without conscious awareness is a topic of considerable importance and ongoing debate. In light of recent theoretical and empirical contributions casting doubt on the early demonstrations of unconscious instrumental conditioning, we revisit the question of its feasibility in two modes of conditioning. In Experiment 1, we used trace conditioning, following a prominent paradigm (Pessiglione et al., 2008) and enhancing its sensitivity. Success in this task requires participants to learn to approach reward-predictive stimuli and avoid punishment-predictive stimuli through monetary reinforcement. All stimuli were rendered unconscious using forward-backward masking. In Experiment 2, we used delay conditioning to shorten the stimulus-outcome delay, retaining the structure of the original task but presenting the stimuli under continuous flash suppression to allow for an overlap of the stimulus, action, and outcome, as well as replacing monetary reinforcement with primary appetitive reinforcement. In both experiments, we found evidence for absence of unconscious instrumental conditioning, showing that participants were unable to learn to adjust their behaviour to approach positive stimuli and avoid negative ones. This result is consistent with evidence that unconscious stimuli fail to bring about long-term behavioural adaptations, and provides empirical evidence to support theoretical proposals that consciousness might be necessary for adaptive behaviour, where selective action is required.

Beyond expectations: the physiological basis of sensory enhancement of satiety.

BACKGROUND/OBJECTIVES: Consumption of high-energy beverages has been implicated as a risk factor for weight gain, yet why nutrients ingested as beverages fail to generate adequate satiety remains unclear. In general, consumers do not expect drinks to be satiating, but drinks generate greater satiety when their sensory characteristics imply they may be filling. These findings challenge traditional bottom-up models of how gut-based satiety signals modify behaviour to suggest that beliefs at the point of ingestion modify gut-based satiety signalling. SUBJECTS/METHODS: Healthy volunteers (n=23) consumed four different beverages, combining an overt sensory manipulation (thin, low sensory (LS) or thicker and more creamy, enhanced sensory (ES)) and covert nutrient manipulation (low energy (LE), 78 kcal; high energy (HE), 267 kcal) on different days. Effects on satiety were assessed through rated appetite and levels of glucose, insulin, pancreatic polypeptide (PP) and cholesystokinin (CCK) recorded periodically over 90 min, and through intake at an ad libitum test lunch. RESULTS: Intake at the test lunch and rated appetite were both altered by both the sensory and nutrient manipulations, with lowest intake and greatest suppression of hunger post-drink in the ESHE condition. Insulin increased more after HE than LE drinks, and after ES than LS drinks, whereas PP levels were higher after ES than LS versions. CCK levels only increased after the ESHE drink. CONCLUSIONS: These data confirm acute sensitivity of satiety after consuming a drink both to the sensory characteristics and nutrient content of the drink, and suggest that this may be, at least in part, due to top-down modulation of release of satiety-related gut hormones.

Does low-energy sweetener consumption affect energy intake and body weight? A systematic review, including meta-analyses, of the evidence from human and animal studies.

By reducing energy density, low-energy sweeteners (LES) might be expected to reduce energy intake (EI) and body weight (BW). To assess the totality of the evidence testing the null hypothesis that LES exposure (versus sugars or unsweetened alternatives) has no effect on EI or BW, we conducted a systematic review of relevant studies in animals and humans consuming LES with ad libitum access to food energy. In 62 of 90 animal studies exposure to LES did not affect or decreased BW. Of 28 reporting increased BW, 19 compared LES with glucose exposure using a specific 'learning' paradigm. Twelve prospective cohort studies in humans reported inconsistent associations between LES use and body mass index (-0.002 kg m(-)(2) per year, 95% confidence interval (CI) -0.009 to 0.005). Meta-analysis of short-term randomized controlled trials (129 comparisons) showed reduced total EI for LES versus sugar-sweetened food or beverage consumption before an ad libitum meal (-94 kcal, 95% CI -122 to -66), with no difference versus water (-2 kcal, 95% CI -30 to 26). This was consistent with EI results from sustained intervention randomized controlled trials (10 comparisons). Meta-analysis of sustained intervention randomized controlled trials (4 weeks to 40 months) showed that consumption of LES versus sugar led to relatively reduced BW (nine comparisons; -1.35 kg, 95% CI -2.28 to -0.42), and a similar relative reduction in BW versus water (three comparisons; -1.24 kg, 95% CI -2.22 to -0.26). Most animal studies did not mimic LES consumption by humans, and reverse causation may influence the results of prospective cohort studies. The preponderance of evidence from all human randomized controlled trials indicates that LES do not increase EI or BW, whether compared with caloric or non-caloric (for example, water) control conditions. Overall, the balance of evidence indicates that use of LES in place of sugar, in children and adults, leads to reduced EI and BW, and possibly also when compared with water.

Caffeine deprivation state modulates coffee consumption but not attentional bias for caffeine-related stimuli.

Previous research has shown that caffeine deprivation state can exert a strong influence on the ability of caffeine to reinforce behaviour. Recent work has also found evidence for an attentional bias in habitual caffeine users. It remains unclear whether deprivation state can influence attentional bias. Here we explored the relationship between caffeine deprivation, attentional bias to caffeine-related stimuli and subsequent caffeine reinforcement measured by consumption of coffee. In three experiments, participants (between-subjects: n=28; within-subjects: n=20, within-subjects: n=20) were preloaded with either caffeine (experiments 1 and 3 : 100 mg; experiment 2 : 150 mg) or placebo, and in experiments 1 and 2 they completed a novel attentional bias task involving pre-attentive word recognition, and in experiment 3 a dot-probe task. In experiments 2 and 3, this was followed by a test of coffee consumption. Greater recognition for caffeine-related words (experiments 1 and 2) and faster reaction times to probes replacing caffeine-related rather than control stimuli (experiment 3) confirmed caffeine-related attentional biases, but in no case was this affected by manipulation of caffeine-deprivation state. Participants in a deprived versus nondeprived state, however, experienced increases in drowsiness and headaches (experiment 2) and reduced alertness (experiment 3). Further, coffee consumption was greatest when participants were caffeine-deprived than when they were nondeprived. Findings are discussed in relation to prevailing theories of drug addiction.

Effects of manipulated palatability on appetite depend on restraint and disinhibition scores from the Three-Factor Eating Questionnaire.

OBJECTIVES: The present study evaluated the effects of dietary restraint on short-term appetite in response to manipulated palatability. DESIGN: The effects of palatability on appetite during a lunchtime meal were assessed by contrasting intake of a bland and palatable version of a simple food (within subject). To test how responses to palatability varied with restraint, these meals consumed by women were classified according to restraint (R) and disinhibition (D) scores from the Three-Factor Eating Questionnaire (TFEQ) as high R/high D (HR-HD), high R/low D (HR-LD), low R/high D (LR-HD) and low R/low D (LR-LD). SUBJECTS: A total of 40 normal-weight women subdivided into four groups based on TFEQ scores. MEASUREMENTS: The overall intake, appetite and hedonic ratings before, during and after the meal. RESULTS: All groups ate similar amounts of the bland food, but the LR-HD group ate significantly more of the palatable version than the other groups, whereas HR-LD did not increase intake in response to palatability. Hunger increased on tasting the palatable food in all but the HR-LD group, and this group ended both meals more hungry/less full than the others. CONCLUSION: Women classified as HR-LD were unresponsive to manipulated palatability, whereas those classified as LR-HD were over-responsive. These findings imply that some individuals are prone to over-respond to palatability and so are at greater risk of developing obesity, whereas others are able to resist the effects of palatability and so successfully self-restrict their food intake. Implications for obesity are discussed.

Failure to reduce short-term appetite following alcohol is independent of beliefs about the presence of alcohol.

Previously, it has been reported that energy consumed as alcohol prior to lunch does not result in subsequent reductions in voluntary food intake, and in some situations alcohol can increase subsequent appetite. The present study extends these findings by examining the effects of beliefs about alcohol content. Eighteen unrestrained men ate lunch 20 min after a preload of either water, an alcoholic beer or a non-alcoholic beer matched for energy content. Food intake was significantly less following the non-alcoholic beer than after alcohol or water, but when preload energy was included subjects had a higher overall energy intake on the day they consumed alcohol compared with both water and no-alcohol conditions. There were no significant differences in hunger or fullness ratings following the three drinks before or after the test meal, but the specific relationship between rated hunger and intake within the test meal was altered by the drink manipulation. The rate at which hunger decreased, and fullness increased, was slower after alcohol than after water or no-alcohol. The drinks did not alter the pleasantness of the test meal or increase hunger at the start of eating. When contrasted with previous work, these data confirm that alcohol consumed before lunch fails to reduce subsequent food intake, but also suggests that changes in rated appetite are influenced by beliefs about alcohol content.

No evidence for latent learning of liking for flavours conditioned by caffeine.

RATIONALE: The ability of caffeine to condition liking for flavours depends on the caffeine deprivation status of subjects; however, it is not known if a latent liking for a flavour can be acquired in an undeprived state, which subsequently emerges when consumers are caffeine deprived. OBJECTIVES: To determine if exposure of undeprived caffeine consumers to a novel drink containing caffeine leads to increased liking for this drink when they are subsequently tested when caffeine-deprived. METHODS: In a double-blind placebo controlled study, four groups of 13 moderate caffeine consumers evaluated a novel flavoured drink on 5 days. The test group consumed this drink with 100 mg caffeine when undeprived on days 1-4, and in a deprived state on day 5. Three control groups had the same conditions on all 5 days, with an undeprived group receiving the caffeinated drink, and two deprived groups receiving the drink with caffeine or placebo. RESULTS: The pleasantness of the drink did not change over the 4 training days in the test group, and did not alter when this group was tested when caffeine-deprived. At no stage did these ratings differ between the test and undeprived control groups. Pleasantness increased significantly over the 5 days in the deprived group who received caffeine, and decreased in the deprived group who received placebo. CONCLUSIONS: These results suggest that repeated pairing of a novel flavour with the effects of caffeine in subjects who are not caffeine deprived does not lead to an emergent liking for that flavour when subsequently tested caffeine-deprived. However, the pleasantness of the same caffeinated drink increased if it was consumed when caffeine deprived.

The actual, but not labelled, fat content of a soup preload alters short-term appetite in healthy men.

The effects of the actual and labelled fat content of a soup preload on appetite at a test meal 30 min later were assessed in 16 healthy men. Each participant ate lunch on four occasions, combining two levels of fat energy (Low, 265 kJ or High, 1510 kJ) and two types of label (Low-fat or High-fat), presented as fictitious soup brand names. Preliminary work established that the Low-fat labels produced an expectation of reduced fat content and lower anticipated hedonic ratings, whereas the High-fat labels generated expectations of a high-fat content and above average hedonic ratings. These expectancies were confirmed in the main experiment, with the soups labelled as high fat rated as both more pleasant and creamy than those labelled low-fat, independent of actual fat content. However, intake at the test meal was unaffected by the preload label, but instead reflected the actual fat (hence, energy) content of the soup, with significantly lower food intake after the high-fat soup regardless of the food label. Rated hunger was lower, and fullness higher, at the start of the meal after the high-fat preloads regardless of how they were labelled, while the pattern of appetite change during the test meal was unaffected by preload. These results suggest that realistic food labels can modify the immediate experience of a consumed food, but do not alter appetite 30 min later in healthy men.

Effects of test-meal palatability on compensatory eating following disguised fat and carbohydrate preloads.

OBJECTIVE: To test whether the palatability of a test meal altered compensatory eating following disguised high-energy fat and carbohydrate preloads. DESIGN: Effects of preload energy (low, 265 kJ, or high, 1510 kJ) and test-meal palatability (bland or palatable) were contrasted within-subjects, with a between-subjects contrast of fat and carbohydrate preloads. SUBJECTS: Twenty-four healthy, normal men (age 23.6+/-1.0 y, (body mass index) BMI 21.3+/-0.5). MEASUREMENTS: Microstructural analysis of test meal intake and rated appetite in the four test conditions, plus diary-based weighed intake analysis of energy intake post-lunch. RESULTS: Subjects ate significantly less at lunch after disguised high-energy fat or carbohydrate preloads relative to the low-energy preload, and ate significantly more of the palatable than bland lunch in all conditions. The reduction in eating following the high-energy preload was significantly less in the palatable condition. Intake post-lunch did not differ between conditions, and overall subjects had higher daily energy intake on the days they consumed the high-energy preloads. Rated hunger was significantly less 30 min after the high- than low-energy preloads, but increased on tasting the palatable food in all conditions. The high-energy preloads suppressed appetite immediately post-lunch. No differences between fat and carbohydrate were found on any measure. CONCLUSIONS: Manipulation of the palatability of a test meal modified the ability to respond to disguised high-energy preloads, with over-consumption most evident when disguised high-energy preloads were followed by a palatable food. Subsequent voluntary intake compensated for over-consumption of the palatable lunch, but not the high-energy preload.

Acquisition and extinction of flavour preferences conditioned by caffeine in humans.

Previous research has shown that moderate caffeine users develop a liking for the flavour of a novel caffeinated drink only if they experience this drink in a caffeine-deprived state. This study tested how sensitive these conditioned-flavour preferences are to subsequent changes in deprivation state and the continued presence or absence of caffeine. Thirty-six moderate caffeine consumers were given 4 training days during which they evaluated a novel flavoured caffeinated drink consumed mid-morning after 12 h caffeine deprivation. Subjects were then divided into four groups depending on whether or not they remained caffeine-deprived and whether the test drink continued to contain caffeine. They then re-evaluated the novel drink over a further 4 test days. As expected, liking for the test drink increased across the 4 training days, and this increased liking was maintained across the 4 test days in the group who continued to receive the caffeinated version of the drink in a caffeine-deprived state. Liking decreased in the test phase in the caffeine-deprived group who no longer received caffeine (extinction). It is surprising that both groups who were tested in a non-deprived state showed a marked decrease in liking on all 4 test days relative to the last training day. This implies that conditioned-flavour preferences may not be expressed in the absence of the relevant motivational state (caffeine deprivation). Together, these data suggest that flavour preferences conditioned by caffeine are very sensitive to changes in the contingent relationship between deprivation state and caffeine content of the drink.

Expression of flavour preferences conditioned by caffeine is dependent on caffeine deprivation state.

RATIONALE: The acquisition of a caffeine conditioned flavour preference depends on the caffeine deprivation status of subjects during conditioning. It is not known if the expression of an established flavour preference is also state-dependent. OBJECTIVES: To determine if the expression of a flavour preference conditioned by caffeine is dependent on the level of deprivation at the time of testing. METHODS: In a double-blind placebo controlled study, 44 subjects were given 4 days exposure to a novel flavoured drink following overnight abstinence from caffeine. Half the subjects received caffeine (100 mg) in the drink, while the remainder had placebo (maltodextrin, 100 mg). Subjects rated the pleasantness of the drink each time. On a fifth (test) day, the subjects were given additional caffeine (100 mg) or placebo 2 h before consuming and rating the pleasantness of the drink. RESULTS: Pleasantness ratings for the novel drink increased over the 4 conditioning days in subjects receiving caffeine, but decreased in those given placebo. On day 5, subjects who were trained and tested in the same caffeine deprivation state expressed pleasantness ratings similar to those for the final training day. In contrast, subjects who were trained and tested in different states expressed pleasantness ratings that were significantly different from those of the final training day. CONCLUSIONS: These results suggest that the expression of caffeine conditioned flavour preferences are acutely sensitive to current motivational state, and a number of possible explanations are discussed.

Rating changes over the course of meals: what do they tell us about motivation to eat?

Detailed analysis of the pattern of change in rated appetite within a meal have proved a useful technique through which to explore appetite control. Variability in individual ratings, and technical difficulties in achieving ratings at equivalent stages of a meal, have lead to the use of curve-fitting techniques to model changes in rated appetite across a meal. These changes could best be described by a quadratic function, in which the three parameters (intercept, linear and quadratic coefficients) represented distinct influences on meal size. In normal subjects, manipulations of palatability and opioid receptor blockade and preloads of alcohol all modified the linear component of this function only, while preloading with maltodextrin reduced appetite at the start of eating (the intercept) but not the pattern of change in ratings within that meal. Thus the linear coefficient appears to measure the degree of stimulation of appetite by the sensory characteristics of the food, while the intercept reflects baseline appetite at the start of a meal. These results suggest that microstructural analyses of rating changes allow some dissociation of the factors underlying motivation to eat, and provide a novel methodology for future experimentation.

Individual differences in the use of pleasantness and palatability ratings.

Previously we have suggested that individual subjects interpret the phrase "palatable food" in different ways. To test the consistency of the use of this term, 50 male volunteers consumed a simple lunch on two occasions, once with a more palatable food and once with a more bland version. Ratings of hunger, fullness, pleasantness and palatability were completed at the start and end of each meal. Overall, subjects ate slightly more of the palatable food. The difference in intake between conditions correlated with differences in pleasantness and palatability but not hunger at the start of eating. Rated pleasantness of the food declined between the start and end of the meal in most subjects, but changes in palatability were more variable, with a significant minority showing no change across the meal. Detailed analysis confirmed the existence of two populations of responses, one where pleasantness and palatability were synonymous, and one where palatability was rated as a constant property of a food. The existence of two interpretations of the term palatability in common English usage cautions against the use of this term as a reliable means of evaluating hedonic responses to foods in appetite studies.

Alcohol and the appetizer effect.

In order to evaluate the effects of alcohol on appetite, 12 unrestrained and 10 restrained men ate lunch 20 min after consuming preloads consisting of water, an alcoholic fruit juice (alcohol) and a non-alcoholic fruit juice (juice). The unrestrained men ate significantly less after the juice preload, and ate most after alcohol. Intake was not altered significantly in the restrained men. However, both the alcohol and juice preloads reduced rated hunger and increased fullness, relative to the water control, in both restrained and unrestrained men. When the relationship between rated appetite and intake within the test meal was modelled mathematically, it was found that hunger increased more during the initial stages of the test meal in the unrestrained men who had consumed alcohol than in any other condition. No such effects were seen in the restrained subjects. Overall, these results suggest that alcohol has a complex action on appetite, which includes some form of appetite stimulation, and this may explain the excess energy intake reported previously in moderate alcohol consumers.

Maltodextrin preloads reduce food intake without altering the appetiser effect.

The effects of consumption of a soup preload with added maltodextrin, relative to a no-maltodextrin control soup matched for sensory properties, on intake and the pattern of changes in rated hunger and fullness during lunch were investigated in 24 male volunteers. Preloads were consumed 30 min before lunch and condition-order counterbalanced. Intake at lunch was reduced significantly by 77 g (407 kJ) after the maltodextrin preload, and this reduced intake was associated with a significant reduction in eating rate but not meal duration. Hunger ratings were significantly lower, and fullness ratings significantly higher, at the start of lunch after the maltodextrin compared with control preload. However, the pattern of changes in subjective appetite once eating had started (assessed by analyzing best-fit quadratic functions between rated appetite and actual intake) did not differ between preloads. Neither the rated pleasantness of the lunch food at the start of the test meal nor the pattern of change in pleasantness across the meal differed between preloads. These results imply that the effect of maltodextrin preloads on appetite is to reduce the general desire to eat, and possible mechanisms for this effect are discussed.

Conditioned flavour preference negatively reinforced by caffeine in human volunteers.

This study examined whether 100 mg caffeine could reinforce preference for the flavour of a novel drink in moderate caffeine users, both after overnight caffeine abstinence and 2 h after receiving 100 mg caffeine, using a two-stage between-groups procedure with 36 volunteers. In the first stage, liking for a test drink (fruit tea) was assessed at breakfast following overnight caffeine abstinence, with half the subjects receiving caffeine. Liking for the tea increased significantly over four trials for subjects receiving caffeine, and decreased significantly in those without caffeine. These effects were greatest in subjects who rated the drink as highly novel. In stage two, subjects evaluated a second drink (fruit-juice) 2 h after receiving the tea, and again half the subjects received caffeine Those subjects who received caffeine in stage two but not stage one showed a significant increase in liking for the fruit-juice over the 4 test days, whereas subjects who did not receive caffeine at either stage showed a progressive decrease in liking for this drink. In contrast, no significant change in liking for the fruit-juice was seen at stage two for subjects who had received caffeine in stage one, regardless of the presence or absence of caffeine at stage two. Caffeine at breakfast increased ratings of energetic and lively, and energetic ratings also increased following caffeine in the fruit-juice in subjects who had not had caffeine at breakfast. Overall, these data are consistent with a negative reinforcement model of caffeine reinforcement, and demonstrate further the utility of the conditioned flavour preference method for evaluating reinforcing effects of drugs in humans.

Exposure to sweetened solutions enhances the anorectic effect of naloxone but not d-fenfluramine.

The effects of prolonged exposure of rats to sweetened caloric (sucrose) and non-caloric (saccharin) solutions on subsequent sensitivity to the anorectic effects of naloxone and d-fenfluramine were investigated in a series of experiments. In Experiment 1, rats given 18 days exposure to 10% sucrose showed greater sensitivity to the anorectic effects of naloxone (0.125-1.0 mg/kg, IP) in a separate feeding test, than did controls or rats exposed to 0.2% saccharin. This effect was replicated in Experiment 2, and here rats exposed to a palatable quinine-sucrose solution that was less preferred than saccharin also showed an enhanced sensitivity to naloxone, similar to that seen in the group exposed to sucrose alone. In Experiment 3, prior exposure to sucrose, quinine-sucrose or saccharin had no effect on the anorectic effects of dexfenfluramine (1.0 and 2.0 mg/kg), while the effect of naloxone (1.0 mg/kg SC) was enhanced by exposure to the two sucrose solutions. All sucrose-exposed rats gained more weight than did control or saccharin-exposed rats. These data suggest that the consumption of palatable calorie-containing solutions selectively alters sensitivity to naloxone, and a number of possible explanations are discussed.

Independent effects of palatability and within-meal pauses on intake and appetite ratings in human volunteers.

The effects of the introduction of timed pauses within meals and palatability on food intake and changes in rated appetite during a meal were assessed in three experiments in which volunteers ate a lunch of pasta with a tomato sauce. Eating was monitored using a disguised electronic balance attached to a micro-computer, which also allowed the introduction of timed pauses within meals. In the first experiment, 16 subjects were tested with both a bland and palatable food (with 0.27% oregano), with eating uninterrupted or with pauses after every 50 g consumed during which appetite ratings were completed. Both the addition of oregano and the introduction of regular within-meal pauses enhanced overall intake. Rated hunger increased in the early stages of eating the palatable food in the interrupted condition, and then declined, whereas hunger declined throughout with the bland food. Similarly, the linear function relating intake to time in the uninterrupted condition was greater with the palatable food. In the second experiment, nine subjects ate the palatable food with no pauses within meals, with 30-second pauses with appetite ratings or with 30-second pauses in a non-appetite task. Intake was greater in both pause conditions than when eating was uninterrupted. In Experiment 3, the effect of pause duration was investigated in a further 16 subjects, with either no pause or a pause of 5, 30 or 60 seconds. Subjects ate more in all pause conditions than with no pauses, while ratings of hunger and fullness suggested that subjects were less satisfied at the end of the meal with longer pauses. These data confirm previous work which suggests that palatability exerts its effect by stimulating appetite and eating rate, but also suggest that the introduction of pauses within meals enhances intake as well, contradicting the idea that pausing within meals should reduce intake by allowing more time for post-ingestive satiety to develop.