Psychiatry Research 220 (2014) 960–969
Contents lists available at ScienceDirect
Psychiatry Research journal homepage: www.elsevier.com/locate/psychres
Tactile mental body parts representation in obesity Federica Scarpina a,n, Gianluca Castelnuovo a,b, Enrico Molinari a,b a b
Psychology Research Laboratory, IRCCS Istituto Auxologico Italiano, Ospedale San Giuseppe, Piancavallo (VCO), Italy Department of Psychology, Università Cattolica del Sacro Cuore, Milan, Italy
art ic l e i nf o
a b s t r a c t
Article history: Received 12 February 2014 Received in revised form 8 August 2014 Accepted 10 August 2014 Available online 6 September 2014
Obese people's distortions in visually-based mental body-parts representations have been reported in previous studies, but other sensory modalities have largely been neglected. In the present study, we investigated possible differences in tactilely-based body-parts representation between an obese and a healthy-weight group; additionally we explore the possible relationship between the tactile- and the visually-based body representation. Participants were asked to estimate the distance between two tactile stimuli that were simultaneously administered on the arm or on the abdomen, in the absence of visual input. The visually-based body-parts representation was investigated by a visual imagery method in which subjects were instructed to compare the horizontal extension of body part pairs. According to the results, the obese participants overestimated the size of the tactilely-perceived distances more than the healthy-weight group when the arm, and not the abdomen, was stimulated. Moreover, they reported a lower level of accuracy than did the healthy-weight group when estimating horizontal distances relative to their bodies, conﬁrming an inappropriate visually-based mental body representation. Our results imply that body representation disturbance in obese people is not limited to the visual mental domain, but it spreads to the tactilely perceived distances. The inaccuracy was not a generalized tendency but was body-part related. & 2014 Elsevier Ireland Ltd. All rights reserved.
Keywords: Obesity Body representation Somatosensory input Size perception Mental imagery
1. Introduction The body may be considered the object about which we constantly receive information, from vision, touch, and proprioception, and from the vestibular and the interceptive systems (De Vignemont, 2011). All of these different sources of information interact with each other to build up our “body representation” (De Vignemont, 2010; Serino and Haggard, 2010). There is a growing consensus that there are (at least) two distinct types of body representation, the body schema and the body image (De Vignemont, 2010). The body schema (Head and Holmes, 1911; Paillard, 1999; Gallagher, 2005) is responsible for the construction of a dynamic representation of one's own body (Dijkerman and De Haan, 2007; Sedda and Scarpina, 2012), which consists of sensorimotor body representations that guide actions (De Vignemont, 2010). The body image (Head and Holmes, 1911; Paillard, 1999; Gallagher, 2005), on the other hand, includes all the representations about the body that are not used for action, whether they are perceptual, conceptual, or emotional (De Vignemont, 2010). These
n Correspondence to: Istituto Auxologico Italiano – IRCCS Ospedale S. Giuseppe, Via Cadorna 90 – 28824 Piancavallo (Oggebbio – VCO), Italy. E-mail address: [email protected]
http://dx.doi.org/10.1016/j.psychres.2014.08.020 0165-1781/& 2014 Elsevier Ireland Ltd. All rights reserved.
representations can be updated selectively (De Vignemont and Farnè, 2010); they also inﬂuence each other (Dijkerman and De Haan, 2007) but still further research is needed to clarify where these two representations cross-talk in the brain (Sedda and Scarpina, 2012). Their reciprocal inﬂuence would depend on the task that subjects are required to solve and the modalities (tactile or visual, action- or perception-related) to perform it (Cardinali et al., 2011). Not only perceptual (visually or tactilely perceived) dimensions of body parts or whole part sizes are shared between the two representations, but also knowledge, beliefs, and attitudes related to the body (De Vignemont et al., 2005; Gallagher, 2005; Dijkerman and De Haan, 2007; Longo et al., 2010) enabling the construction of an integrated sense of one's own body in a dynamic environment (Dijkerman and De Haan, 2007). Indeed emotions about the body (in which one's body is the object of the emotions) are frequently expressed in terms not only of whole body size but also of certain body parts' sizes (Longo et al., 2010); thus when people are asked to respond about speciﬁc parts of their bodies, feelings and cognitive concepts relative to those body parts are activated (Shontz, 1969). In eating disorders, the visual component is traditionally the most investigated sensory source relative to the mental whole-body representation (Glucksman and Hirsch, 1969; Slade and Russell, 1973; Garner et al., 1976; Wingate and Christie, 1978; Kalliopuska, 1982; Bell et al., 1986; Collins et al., 1987; Probst et al., 1992;
F. Scarpina et al. / Psychiatry Research 220 (2014) 960–969
Valtolina, 1998; Docteur et al., 2010), assuming that the distortion of perceptual dimension is of the same magnitude for all parts of the body (Slade and Russell, 1973), and relative to the mental representation of speciﬁc body parts (Gardner et al., 1987; Yamokoski, 1975; Pearlson et al., 1981; Fisher, 1986; Cafri and Thompson, 2004). Three different patterns relative to the perceptual estimation of body image in obese individuals have been reported (Schwartz and Brownell, 2004): they seem to (1) overestimate (Garner et al., 1976; Collins et al., 1987; Gardner et al., 1989; Docteur et al., 2010), (2) underestimate (Bell et al., 1986; Valtolina, 1998) or (3) be accurate (Schwartz and Brownell, 2004) regarding whole body-size estimation. About body-parts estimation, it has also been reported that obese people are generally less accurate than people of healthyweight (Fisher, 1986; Cafri and Thompson, 2004); speciﬁcally, they generally showed a trend of overestimation (Yamokoski, 1975; Gardner et al., 1987). On the other hand, similarities in obese and healthy-weight people's performance in size judgments of separate body parts have also been observed (Gardner et al., 1987), and differences have appeared to be related to gender (Pearlson et al., 1981). There is a growing interest about the investigation into how the body or body parts are perceived in obesity (Schwartz and Brownell, 2004). Obesity cannot be deﬁned merely as a medical– physiological phenomenon, but its manifestations extend also to the psychological and cognitive domain (Kreitler and Chemerinsky, 1990; Friedman and Brownell, 1995). About the former, even though the causal relationship among obesity, mood disorders, and general medical illness is far from being completely understood (see McElroy et al. (2004) for a review), in obese patients a negative body image appeared to be related to signiﬁcant psychological problems, including depression, low self-esteem and a nonfunctional quality of life (Friedman and Brownell, 1995; Friedman et al., 2002). Lo Coco et al. (2014) suggested that obesity is strictly correlated with body image dissatisfaction, that has been indicated as a potential mediator of the relationship between dysphoric psychological states and obesity (Friedman and Brownell, 1995; Legenbauer et al., 2011; Nicoli and Junior, 2011; Lo Coco et al., 2014). Cognitive manifestations of obesity include transformations in the form and function of the mental body representation (Kreitler and Chemerinsky, 1990). It was reported that the overestimation of shape and weight appeared to be not only related to the frequency of binge-eating episodes, but also strictly connected with general psychological distress (Grilo et al., 2012). Obese individuals that overestimate or distort the size of their body are more dissatisﬁed and preoccupied with their appearance and tend to avoid more social interactions because of their appearance than healthy-weight individuals do (Gardner et al., 1987; Tiggeman and Rothblum, 1988; Cash, 1990); moreover, most of them were likely to drop out of treatment (Collins et al., 1987). On the other hand, one effect of clinical treatment for obesity was reported to be a decrease in the overestimation pattern of body size by obese women: this phenomenon was linked to increased self-efﬁcacy and a positive self-image (Bell et al., 1986; Valtolina, 1998). In previous studies, sensory modalities other than vision seemed to have been mostly neglected in the assessment of body parts representation in obesity. However, mental body representation is constructed on the basis of multiple sources (De Vignemont, 2010; Serino and Haggard, 2010): not only visual and tactile sensations, but also cues from other sensory modalities, such as proprioceptive, auditory, and vestibular cues, do contribute to complete this representation (Serino and Haggard, 2010). Thus in the present study, we aimed to test obese and healthy-weight subjects in a task in which the subjects infer a perceptual dimension of two body-part sizes based on tactile judgment; moreover, we sought to investigate the relationship between this judgment and the visual dimension of body image. Based on
previous results about distorted body image representation in obesity, we hypothesized that obese people would show a distortion of the tactile distances perceived on their body, conveying an inaccurate body-parts representation. Based on the lack of previous research in this speciﬁc domain, we could not hypothesize a-priori regarding the question if obese people would tend to overestimate or underestimate the tactilely perceived distance; however, we reasoned that the distortion would be associated with an inaccurate visually-based mental body representation. In order to investigate our hypothesis, we borrowed the experimental methodology from Keizer et al.'s (2011) study in which the authors investigated mental body-parts' representation in anorexia nervosa patients through a tactilely-based estimation task. Participants were required to estimate the distance between two tactile stimuli presented simultaneously on body-part surfaces (Keizer et al., 2011; Taylor-Clarke et al., 2004); two body parts were stimulated, the abdomen, as a likely high-concern body part, and the arm, as a likely neutral body part (Keizer et al., 2011). Perceptual judgment depends not only on the neurophysiologic characteristics of the touched body parts, but also on the internal model of its physical size (Serino and Haggard, 2010). Since the task required to refer implicitly to the size of the touched body part (Serino and Haggard, 2010; Longo et al., 2010), the subjective judgment would be directly inﬂuenced by the mental body-part representation (Serino and Haggard, 2010; Spitoni et al., 2010). Keizer et al. (2011) suggested that this task would measure the tactilely-perceived body image, thus enabling the indirect link between the judgment of a distance perceived by touch and the characteristics of mental body-part reconstruction. We also explored the visual aspects of body image by a visual imagery task that allowed us to assess spatial relationships between distances on an individual's body (Smeets et al., 2009). The assumption behind this task is that topological relationships between an object's parts, and even the object's metric information, are preserved in the mental image as they are physically in the object (Kosslyn, 1980; Denis, 2008; Smeets et al., 2009). In this task, we asked participants to mentally assess two distances on their bodies and to decide which one was longer or shorter as quickly as possible (Smeets et al., 2009). The targets were divided into two groups, body parts that are sensitive to preoccupation about size and shape (Molinari, 1995; Smeets et al., 2009) and insensitive body parts. Smeets et al. (2009) suggested that: (1) people who have inappropriate body images reported different levels of accuracy and reaction times with respect to controls and (2) more deeply, body parts that are sensitive to body-shape concerns require more time to be scanned than insensitive body parts do. For its characteristics, this task can be assimilated with the classical methodologies used to assess body representation in cases of eating disorders, such as distorting photographs (Glucksman and Hirsch, 1969; Garner et al., 1976; Collins et al., 1987; Docteur et al., 2010), videos (Probst et al., 1992), silhouette charts (Bell et al., 1986), and drawings (Wingate and Christie, 1978; Kalliopuska, 1982; Valtolina, 1998). However, these techniques would present some critical issues. First, there is an implicit distinction based on the stimulus between the different tasks: the crucial distinction seems rather to be whether the stimulus being compared to the body is a depiction of a body (“depictive” methods) or merely a metric standard (“metric” methods) (Longo and Haggard, 2012). Selecting a silhouette or a photograph that represents the subjects' perceived dimensions of his or her body would require spatial abilities as well as perceptual interpretations and productions (Bell et al., 1986). Indeed people's perceptual judgments of their own bodies are based on external frames that are not body-integrated and that require complex cognitive processes to be applied; subjects have to create a mental image of their body parts and inspect them for size in order to judge if
F. Scarpina et al. / Psychiatry Research 220 (2014) 960–969
the external support matches the mental representation (Keizer et al., 2011). Moreover it was suggest that selecting an image that reﬂects how overweight participants effectively feel represents a measure of body satisfaction and not a measure of mental body representation (Smeets et al., 2009). From an experimental point of view, if participants have an idea about the aim of the task or the hypothesis being test, they would implicitly apply a strategic and conscious alteration of the distance estimations (demand effect, Smeets et al., 2009). The distance comparison task was created with the aim to avoid these possible confounding effects and to determine speciﬁcally the visually-based mental body representation (Smeets et al., 2009).
were blindfolded for the duration of the task. The experimenter simultaneously, lightly pressed the two pointers of a calliper on the participants' skin. Participants were asked to estimate the distance between the two tactile stimuli by varying the separation between the thumb and the index ﬁnger of the left hand, and then the experimenter measured this separation with a ruler. Two body parts were assessed: (1) the ventral side of the right forearm was stimulated along the longitudinal (vertical) axis representing a distal body part; (2) the upper part of the abdomen was stimulated along the horizontal axis, representing a proximal body area. According to Keizer et al. (2011), the two body parts are differently related to rate of satisfaction and concern about the process of slimming down. Additionally Sarwer et al. (1998) reported that the vast majority of obese reporting the greatest dissatisfaction with their waist or abdomen. The distance between the two pointers was set at 5 cm, 6 cm and 7 cm. The blocks of judgments for the forearm and the abdomen were presented with an AB–BA design. Overall, 84 trials for each participant were presented: for each block, seven repetitions for each cell of the experimental design 2 3 (body part measure) were presented in random order.
2. Methods 2.1. Participants The current study was approved by the ethical committees of the IRCCS Istituto Auxologico Italiano, and it was performed in compliance with Declaration of Helsinki's ethical principles (World Medical Association, 1991). Twenty-one (11 females; 10 males) obese patients and 21 (12 females; 16 males) healthy-weight participants took part in this experiment. All participants were right-handed and free from scar tissue on their right arms and their abdomens. All obese patients were recruited during the ﬁrst weeks of a rehabilitation recovery in the IRCCS Istituto Auxologico Italiano – Ospedale San Giuseppe. They were seen for screening interviews for admission to the study. Exclusion criteria for the study were: (1) severe psychiatric disturbance diagnosed by DSM-IV-TR criteria (except for BED, DSM-IV-TR criteria) and (2) any concurrent medical condition not related to obesity. Structured Clinical Interview for DSM-IV-TR Disorders (SCID) I and II were used as screening tools for psychiatric disorders and were administered by an independent clinical psychologist as part of his work. The participants of the present study were people who had been hospitalized in order to lose weight and at the start of their rehabilitation. They knew they would receive information about diet and healthy life styles and that they would be engaged in physical activities and psychological support. We would like to emphasize that the people participating in the present study seemed to be aware of their obesity. The healthy-weight group was recruited outside the clinical institute; the exclusion criterion was a body mass index over 24.9. The possible presence of any eating disorders, speciﬁcally anorexia nervosa and bulimia nervosa, were excluded by the SCOFF questionnaire (Morgan et al., 1999; Pannocchia et al., 2011): a threshold of two or more positive answers to all ﬁve questions would suggest a possible disease or disorder. All healthy-weight participants reported a score lower than or equal to 2 points. The mean age was 45 years (SD ¼10) for the obese group and 29 years (SD ¼ 4) for the control group: the obese participants were signiﬁcantly older than the healthy-weight participants [t(47) ¼ 6.88; p o0.001]. The obese group's mean education (mean¼ 11 years, SD ¼4) was signiﬁcantly lower than was control group's (mean ¼17 years, SD ¼2) [t(47) ¼ 6.03; p o 0.001]. At the MMSE – Mini Mental State Examination (Folstein et al., 1975) the obese group reported a signiﬁcantly lower mean score, at 28.8 (SD ¼1.69), than the control group, at 29.57 (SD¼ 0.69) [t(47) ¼2.61; p ¼ 0.03]. All participants reported a score over the cut-off of 24 points. As expected, the two groups differed signiﬁcantly in body mass index (BMI) with a mean of 40.78 (SD ¼ 8.09; range 30–58) for the obese group and a mean of 21.84 (SD ¼1.69; range 19–25) [t(46) ¼ 10.15; p o 0.001] for the healthy-weight participants. 2.2. Materials and procedure 2.2.1. Likert scale Before starting with the tactile size-estimation task experiment, the participants were asked to rate how dissatisﬁed they were with the physical aspects of their arms and abdomens on a seven-point scale (“What's the rate of dissatisfaction for the physical aspects about your arm/ abdomen?”). Subjects were informed that 1 indicated “very satisﬁed” and 7 indicated “dissatisﬁed.” After the tactile sizeestimation task, participants were asked to rate how concerned they were with their arm and abdomen not slimming down on a seven-point scale (“What's the rate of your concerning bout the fact your arm/abdomen not slimming down?”). Participants were informed that 1 indicated “not worried” and 7 indicated “highly worried.” 2.2.2. Tactile size estimation task This tactile size-estimation task was an adapted version from Keizer et al. (2011) and speciﬁcally investigated tactilely-perceived body image. Participants
2.2.3. Distance comparison task This image-scanning task, an adapted version from Smeets et al. (2009), speciﬁcally investigated the visually-perceived body image; participants were asked to compare two of their own body horizontal widths. Preliminarily, the participants were asked to close their eyes and then, following the experimenter's instructions, they were invited to construct a visual, mental image of their body. They were asked to imagine themselves standing upright, with feet together, in front of a full-length mirror. They were then invited to imagine inspecting and focusing on the actual distance between the following body parts: ears, shoulders, armpits, elbows, waists, hips, thighs, and knees. For example, they were asked to imagine scanning the distance between the right ear and the left ear. The experimenter verbally indicated the margins of the body distances to be scanned. Afterward, the participants were presented to the computer task. The task was coded in OpenSesame 0.27.2 (Mathôt et al., 2012). In each trial, two word pairs consisting of two identical body parts (representing the left and the right parts) were shown consecutively. Word pairs consisting of the body parts waist, hips, and thighs were considered sensitive pairs while word pairs with ears, shoulders, armpits, elbows, and knees represented insensitive pairs. In each trial, two word pairs were consecutively shown. A total of 28 word-pair combinations were created. Each combination was presented two times in the order AB (for example ﬁrst waist–waist and then hip–hip) and two times in the order BA (ﬁrst hip–hip and then waist–waist). Overall 112 word-pair combinations were presented in two different blocks, in counterbalance order. Participants were asked to look at the monitor's screen. Each trial started with a central cross of 2000 ms duration, after which the ﬁrst word pair appeared for 500 ms. Then a blank screen of 250 ms appeared followed by the second word pair that was displayed for 500 ms. Finally a blank screen appeared that lasted until a participant's response was recorded. The participants were required to indicate if the ﬁrst presented word pair reﬂected a larger or a shorter horizontal distance on their own body than the second presented word pair. They were invited to be accurate and to answer as quickly as possible. Additionally, they were invited to press a different key on the keyboard when they had not read one or both word pairs. The stimuli's presentation and order of the key answer labels were randomized across participants. Answer accuracy and reaction times (RTs) were recorded. After the completion of the experiment, the relevant distances on each participant's body were measured. 2.3. Analyses 2.3.1. Likert scales Regarding the Likert scale relative to the rate of satisfaction, an independent t-test between the two groups (the obese versus the healthy-weight group) was conducted to ﬁnd the scores relative to the arm and to the abdomen. For each group, a paired sample t-test was also conducted to explore possible differences between the two explored body parts (the arm and the abdomen). The same procedure was used with the Likert scale relative to the rate of concern about the weight loss of the arm and the abdomen. 2.3.2. Tactile size estimation task Overall, 1.5% of trials relative to the obese group and 2.08% of the trials relative to the healthy-weight group were excluded from the analyses because the participants did not perceive the tactile input. The difference between the estimated distance and the real distance was measured for each trial, with the distance representing the error. A negative error indicated an underestimation of the distance; a positive error, an overestimation. Trials in which the error was out of the range of two standard deviations of the group's mean were excluded from the analysis, which included an overall 3.11% of trials relative to the obese group and 2.67% of trials relative to the healthyweight group.
F. Scarpina et al. / Psychiatry Research 220 (2014) 960–969 A mixed, repeated measures analysis of variance was performed with the variable of group (obese versus healthy-weight group) as the between-subjects factor and the variables of body part (abdomen versus arm) and measure (5 cm, 6 cm, and 7 cm) as the within-subjects factors. Regarding the arm and the abdomen independently, the errors for the three different measures (5 cm, 6 cm, and 7 cm) were collapsed together. Referring to obese group, an independent t-test was conducted with the variables of gender (males versus females), in order to verify a possible effect of gender on the performance relative to the arm and the abdomen. Moreover, Pearson product– moment correlation coefﬁcients were calculated in order to explore a possible relationship between the judgment error relative to the arm and the abdomen independently with the BMI value, the age variable and with the rates of dissatisfaction and concerning the weight loss of the arm and the abdomen, as measured by Likert scales. The same set of analyses was conducted for the healthy group.
2.3.3. Distance comparison task Overall 11.43% of the trials relative to the obese group and 9.9% of the trials relative to the healthy-weight group were excluded from the analyses because the participants did not give an answer. Moreover overall 4.2% of the trials relative to the obese group and 5.26% of the trials relative to the healthy-weight group were excluded since RTs were o 200 ms and 49000 ms. For each participants, the accuracy (i.e. correctly judged distance differences) of each response was determined by inspecting the real horizontal widths of body parts compared with the participant's answer. The accuracy was expressed as percentage of accurate responses on the amount of valid trials. A mixed, repeated measures analysis of variance was performed with the variable of group (obese versus healthy-weight group) as the between-subjects factor and the variable of category (sensitive versus insensitive) as the withinsubjects factor. The accuracy relative to the sensitive and insensitive body parts were collapsed together. Referring to obese group and for the healthy group independently, an independent t-test was conducted with the variables of gender (males versus females), in order to verify a possible effect on the accuracy. Moreover, Pearson product–moment correlation coefﬁcients were calculated in order to explore a possible relationship between the accuracy with the BMI value and the age variable. The same set of analyses was conducted for the RTs.
2.3.4. Relation between different modalities representation In order to verify possible relationship between the tactile estimation in tactile size estimation task and the visual judgement of horizontal distances in distance comparison task, Pearson product–moment correlation coefﬁcients were calculated between the tactile errors' average from the ﬁrst task and the accuracy and RTs averages from the second one. The same set of analyses was conducted independently for the obese group and the healthy group.
3. Results 3.1. Likert scales Participants were asked to rate how dissatisﬁed they were about the physical aspects of the arm and the abdomen on a seven-point scale. The Shapiro–Wilks test indicated that the variable of rate for the arm and the abdomen were normally distributed for the healthy group [p 40.67] but not for the obese group [p o0.009]. According to the Mann–Whitney test, the obese group (arm mean ¼5.04, SD ¼1.93; abdomen mean ¼ 6, SD ¼1.58) reported a signiﬁcantly higher rate of dissatisfaction for the arm [U¼107; p o0.001; r ¼0.55] and the abdomen [U¼108; po 0.001; r ¼0.54] than the healthy-weight group (arm mean ¼2.82, SD ¼1.38; abdomen mean ¼3.78, SD ¼1.83); these results was observed also when the planned independent t-test was run [about arm: [t(47) ¼ 4.69; p o0.001; η2 ¼0.31]; about abdomen: [t(47) ¼ 4.33; p o0.001; η2 ¼0.66]. Moreover, according to the result of the Wilcoxon Signed-rank test, there was no signiﬁcant difference between the score for the arm and the abdomen for the obese group [Z¼1.83; p ¼0.06; r ¼0.39], as conﬁrmed also by the planned paired-sample paired sample t-test [t(40) ¼1.74; p ¼0.08; η2 ¼0.13]: this result suggests that the obese participants were equally dissatisﬁed about the arm and the abdomen. However, the healthy-weight group showed higher levels of dissatisfaction with the abdomen than with the arm [t(54) ¼2.21; p¼ 0.03; η2 ¼ 0.15].
Participants were asked to rate how concerned they were about their arm and abdomen not slimming down on a seven-point scale. The Shapiro–Wilks test indicated that the variable of rate for the arm and the abdomen were not normally [po 0.05]. About the arm, the obese group (arm mean ¼4.71, SD¼ 2.12) rated higher levels of concern than did the healthy-weight group (arm mean ¼1.82, SD ¼ 1.46) [U¼ 81; po 0.001; r ¼0.09]. Also about the abdomen, the obese group (abdomen mean ¼ 5.95, SD ¼2.03) rated higher levels of concern than did the healthy-weight group (abdomen mean ¼4.01, SD ¼2.04) [U¼132; p o0.001; r ¼0.48]. Overall these results suggest that obese people were more concerned that these body parts would not slim down than were the healthy-weight participants. Moreover, there was a signiﬁcant difference between the scores for the arms and the abdomens for the obese group [Z¼ 4.18; p o0.001; r ¼0.92] and for the healthy-weight group [Z¼2.46; p¼ 0.014; r ¼0.46], suggesting higher levels of concern about their abdomens not slimming down than their arms. 3.2. Tactile size estimation task We aimed to explore the possible effect of obesity in the judgment of the horizontal distance perceived tactilely compared to healthy-weight participants' perception using two different body parts, the arm and the abdomen. The Shapiro-Wilks test was used to test the assumption of normality of the involved variables: each of the levels of the independent variable (the error) was normally distributed (p 40.11). Although there was no signiﬁcant main effect of group (obese group: mean ¼2.9, SD ¼2.93; healthy-weight control: mean ¼1.98, SD ¼1.8) [F(1,47) ¼2.84; p ¼0.09; η2 ¼0.057], there was a signiﬁcant main effect of body part [F(1,47) ¼11.42; p¼ 0.001; η2 ¼0.19] on the judgment error: the distance on the arm (mean ¼2.83, SD¼ 2.2) was overestimated more than the distance on the abdomen (mean¼ 1.92, SD ¼2.49). More interestingly, a signiﬁcant interaction between group and body part emerged [F(1,47) ¼ 14.24; p o0.001; η2 ¼0.23]. According to the Bonferroni-corrected estimated marginal mean comparisons, a signiﬁcant difference was found in the judgment of the distances tactilely perceived on the arm between the obese group (mean¼4.04, SD ¼2.26) and the healthy-weight group (mean¼1.92, SD ¼1.65) [p o0.001, d ¼1.7] but not relative to the distances perceived on the abdomen (obese group: mean ¼1.75, SD¼ 3.09; healthy-weight group: mean ¼2.05, SD ¼0.92) [p ¼ 0.67, d¼ 0.13]. Moreover, the obese group signiﬁcantly overestimated more the distances of the arm than those on the abdomen [p o0.001; d ¼0.84], while no difference emerged relative to the healthy-weight group [p ¼0.76; d ¼ 0.09] (Fig. 1). These results indicate that both groups showed a similar error in the judgment of the tactilely perceived distances on the abdomen, but the obese group showed a greater overestimation than did the healthyweight group when the tactile inputs were administered on the arms. A signiﬁcant main effect of measure emerged [F(2,94) ¼7.16; p¼ 0.005; η2 ¼0.13]; Bonferroni-corrected, estimated marginal mean comparisons indicate that the distances of 5 cm (mean¼2.57, SD ¼2.43) [p ¼0.018; d ¼0.17] and 6 cm (mean ¼2.41, SD¼ 2.37) [p¼ 0.007; d ¼0.11] were signiﬁcantly more overestimated than were the distances of 7 cm (mean¼ 2.14, SD ¼2.38), while no difference emerged between them [p ¼0.33; d ¼0.06]. This effect did not interact with the variable of group, [F(2, 94) ¼ 0.17; p ¼0.83; η2 ¼0.004] nor with the variable of body part [F(2, 94) ¼0.16; p ¼0.85; η2 ¼0.003]. These results suggest a general tendency to overestimate the tactilely perceived distances; as the distance between the two points decreased, the error became larger. The effect of distance between the two simultaneously
F. Scarpina et al. / Psychiatry Research 220 (2014) 960–969
applied tactile stimuli on the index ﬁnger–thumb separation was not relevant to the aims of the current study, speciﬁcally because it did not interact with the group. However we would suggest that this effect refers to additional cognitive processes applied on the perceptual judgment in order to correct for possible distortions dependent on primary representations of the stimuli (Longo et al., 2010) since the tactile perception of size seemed not to be veridical. Finally, the second order interaction between group, body part, and measure was not signiﬁcant [F (2,94)¼0.094; p¼0.91; η2 ¼0.032]. Considering the obese group, no effect of gender on the tactile estimation error relative to the arm [t(19) ¼ 0.06; p ¼0.95; η2 o 0.001] and the abdomen [t(19) ¼0.54; p ¼0.59; η2 ¼0.01] was found; according the Shapiro–Wilks test was used to test the assumption of normality, the involved variables were normally distributed [p 40.54] (Table 1). The error relative to the arm was not related to BMI [r ¼ 0.005; n ¼20; p ¼0.98] or to Age [r ¼ 0.17; n ¼20; p¼ 0.45]; moreover the estimation of the tactile distance was not related to the rate of dissatisfaction [r¼ 0.3; n ¼21; p ¼0.87] or rate of concern that this body part would not slim down [r ¼0.17; n ¼21; p ¼ 0.45]. The same pattern of results was
found about the abdomen, since no correlation with the BMI [r ¼ 0.15; n¼ 20; p ¼0.51], age [r ¼0.12; n¼ 21; p ¼0.057], rate of dissatisfaction [r ¼ 0.13; n ¼21; p ¼0.57] or rate of concern [r ¼ 0.13; n ¼21; p ¼0.56] emerged from the analyses. Considering the healthy group, no effect of gender on the tactile estimation error relative to the arm [t(26) ¼ 0.096; p ¼0.92; η2 o0.001] and the abdomen [t(26) ¼0.37; p ¼0.71; η2 ¼ 0.004] emerged from the analyses; according the Shapiro–Wilks test was used to test the assumption of normality, the involved variables were normally distributed [p 40.43] (Table 1). The error relative to the arm was not related to BMI [r¼ 0.027; n ¼28; p ¼0.16] or to age [r ¼ 0.76; n ¼28; p ¼0.7]; moreover the estimation of the tactile distance was not related to the rate of dissatisfaction [r ¼ 0.74; n¼ 28; p ¼0.7] or rate of or rate of concern that this body part would not slim down [r ¼ 0.15; n ¼28; p ¼0.79]. The same pattern of results was found about the abdomen, since no correlation with the BMI [r ¼0.56; n ¼28; p ¼0.77], age [r ¼ 0.23; n ¼28; p ¼0.23], rate of dissatisfaction [r ¼0.18; n¼ 28; p ¼0.34] or rate of concern [r ¼ 0.29; n ¼28; p ¼0.88] emerged from the analyses. 3.3. Distance comparison task
Fig. 1. Mean distance estimation in the tactile size estimation task in cm (y axis; positive value: overestimation; negative value: underestimation) by group (dark grey¼ healthy-weight group; light grey¼ obese group) and body part (x axis: arm versus abdomen). Error bars depict the standard error. n Represents signiﬁcant difference at p ¼0.001.
The present analyses had to explore possible effects of obesity in the construction of a visually-based mental representation of obese participants' bodies compared to those of healthy-weight participants. Different body parts were used for the investigation, based on their rate of sensitivity. Regarding Accuracy, the Shapiro–Wilk test indicated that this variable was not normally distributed (p o0.004). The Mann– Whitney test was used to compare the accuracy of the two groups: the obese group (mean¼ 67.33, SD ¼18.04) was less accurate than was the healthy-weight group (mean ¼88.71, SD ¼7.34) [U¼16; po 0.001; r ¼ 0.8]. The Wilcoxon Signed-ranks test indicated that the accuracy for the sensitive parts (mean¼ 74.71, SD ¼15.66) was lower than it was for the insensitive parts (mean¼84.45, SD ¼ 15.66) [Z¼ 5.95; p o0.001; r ¼0.79]. About the accuracy relative to sensitive pars, the Mann–Whitney test revealed that the obese group (mean ¼59.27, SD ¼9.82) was less accurate than was the healthy-weight group (mean ¼86.30, SD ¼6.32) [U¼ 8; p o0.001; r ¼0.82]; also, the obese group (mean¼ 75.4, SD¼ 20.84) was less accurate than was the healthy group (mean ¼91.23, SD ¼7.56) also for the insensitive part [U¼ 53; p o0.001; r ¼0.69]. Additionally, the obese group showed lower level of accuracy for sensitive parts respect to the insensitive parts [Z ¼ 3.84; p o0.001; r ¼0.83]; a similar result was found for the healthy group [U¼ 4.16; po 0.001; r ¼0.78]. Since this results, the planned mixed, repeated
Table 1 For the obese and healthy groups, demographical information and performance at the experimental tasks were reported split for males and females. Obese group
Number of participants Age in years Education in years MMSE score BMI as mass in kg/(height in cm)2
10 40 (7) 11 (4) 28.55 (2.18) 45.56 (8.87)
11 48 (11) 12 (3) 29 (1.27) 35.34 (4.89)
16 30 (4) 17 (2) 29.47 (0.7) 22.73 (1.47)
12 29 (5) 17 (2) 29.72 (0.4) 20.45 (0.89)
Tactile size estimation task Arm: error in cm Abdomen: error in cm
3.74 (2.2) 1.7 (2.3)
4.2 (2.21) 1.7 (3.66)
1.94 (1.44) 1.92 (2.01)
1.88 (1.83) 2.2 (1.75)
Distance comparison task Percentage of accuracy Rts in ms
67.34 (8.12) 3217 (1159)
61.52 (11.2) 2816 (998)
87.43 (6.56) 3335 (1046)
86.47 (6.5) 2813 (571)
F. Scarpina et al. / Psychiatry Research 220 (2014) 960–969
measures analysis of variance was performed. A signiﬁcant difference between the groups emerged [F(1,47) ¼56.69; p o0.001; η2 ¼0.54], since the obese group was less accurate than was the healthy-weight group. Moreover, a signiﬁcant effect of category emerged from the analyses [F(1,47) ¼ 30.15; p o0.001; η2 ¼ 0.39 Greenhouse-Geisser corrected] since the mean accuracy for the judgment of the sensitive parts was lower than it was for the insensitive parts. The interaction between group and category was signiﬁcant [F(1,47) ¼8.16; p o0.005; η2 ¼ 0.15]: the Bonferronicorrected, estimated marginal mean comparisons showed the level of accuracy was signiﬁcantly lower for sensitive body parts than for insensitive ones for the obese participants [po 0.001; d ¼0.99], while the healthy-weight group showed the same level of accuracy [p¼ 0.054; d ¼0.7]. Moreover, the obese participants appeared to be signiﬁcantly less accurate than the control group relative to the sensitive [p ¼0.001; d ¼3.27] and insensitive body parts [p o0.001; d ¼1] (Fig. 2, panel A). Considering the obese group, no effect of gender on accuracy emerged [Shapiro–Wilk test p ¼0.04; U¼37.5; po 0.24; r ¼0.25] (Table 1). A signiﬁcant positive correlation emerged with the BMI [r ¼0.44; n ¼20; p ¼0.49] while a negative relation was found with the age [r ¼ 0.57; n ¼20; p ¼0.006]. Considering the healthy group, no effect of gender on accuracy emerged [Shapiro–Wilk test p¼ 0.08; t(26) ¼0.38; p ¼0.7; η2 ¼ 0.005] (Table 1) neither signiﬁcant relation with the BMI
[r ¼ 0.29; n ¼28; p¼ 0.24], neither with the age [r ¼0.077; n ¼28; p ¼0.69]. About the RTs, the Shapiro–Wilk test indicated that each level of the independent variable was normally distributed (p4 0.066). There was no signiﬁcant difference between the obese group (mean¼2988, SD ¼1061) and the healthy-weight control (mean¼3111, SD¼ 900) [F(1,47) ¼ 0.9; p ¼0.66; η2 ¼0.04]. Interestingly, there was a signiﬁcant difference in relation to the variable category [F(1,47) ¼56.64; p o0.001; η2 ¼ 0.54; Greenhouse-Geisser corrected]: the sensitive parts (mean¼ 3244, SD ¼1018) required more scanning times than did the insensitive parts (mean¼2873, SD¼937). Additionally, a signiﬁcant interaction between group and category emerged [F(1,47)¼5.61; p¼0.022; η2 ¼0.1 GreenhouseGeisser corrected]. According to the Bonferroni-corrected, estimated marginal mean comparisons, the previous result relative to the longer time required for the sensitive parts with respect to the insensitive body parts emerged not only for the obese group (sensitive body parts mean¼3109, SD¼1143; insensitive body parts mean¼2866, SD¼1001) [p¼0.001; d¼0.22] but also for the control group (sensitive body parts mean¼3345, SD¼923; insensitive body parts mean¼ 2878, SD¼905) [po0.001; d¼ 0.51]; however, no difference emerged between the obese and the healthy-weight participants about RTs in relation to the sensitive [p¼ 0.96; d¼0.22] and the insensitive [p¼0.43; d¼ 0.01] word pairs (Fig. 2, panel B).
Fig. 2. About the distance comparison task, mean percentage of accuracy (y axis, part A) and mean RTs (y axis, part b) by group (dark grey¼ healthy-weight group; light grey ¼ obese group) and body part (x axis: sensitive versus insensitive). Error bars depict the standard error. n Represents signiﬁcant difference p ¼ 0.001, instead nn at p ¼ o 0.001.
F. Scarpina et al. / Psychiatry Research 220 (2014) 960–969
Considering the obese group, no effect of gender on RTs emerged [Shapiro–Wilk test p ¼0.04; t(19) ¼ 0.84; p ¼0.4; η2 ¼ 0.03], neither signiﬁcant relation with the BMI [r¼ 0.14; n ¼20; p ¼0.55], neither with the age [r ¼ 0.36; n ¼20; p¼ 0.1]. Considering the healthy group, no effect of gender on RTs emerged [Shapiro–Wilk test p¼ 0.54; t(26) ¼ 1.55; p ¼0.13; η2 ¼0.08], neither signiﬁcant relation with the BMI [r ¼ 0.04; n ¼28; p ¼0.81], neither with the age [r ¼ 0.09; n ¼28; p ¼0.64]. 3.4. Relation between different modalities representation A negative correlation [r(49) ¼ 0.29; p ¼0.043] between the tactile error's mean (mean¼2.37; SD ¼1.91) in the tactile estimation and the accuracy's mean in the Distance Comparison Task (mean¼ 77.05; SD ¼13.93), but not with the RTs' mean (mean¼ 3068; SD ¼ 964) [r(49) ¼0.2; p ¼0.16], emerged from the analyses: the overestimation of the tactile distance was related to lower level of accuracy in the visual estimation of horizontal distances, but not with the time used to explore them. About the obese group, a negative correlation between the error mean (mean ¼2.9; SD ¼2.31) in the tactile estimation and the RTs mean (mean¼ 2988; SD¼ 1061) [r(21) ¼ 0.43; p ¼0.05], but not with the accuracy mean (mean ¼64.02; SD ¼10.2) [r(21) ¼ 0.01; p¼ 0.95] in the distance comparison task, emerged from the analyses, suggesting that the overestimation would be related to lower amount of time used to explore them. This pattern of results was not replicated in the healthy group, since no signiﬁcant correlation emerged between the errors mean (mean ¼1.98; SD ¼1.47) in the tactile estimation and the accuracy mean (mean¼ 87.02; SD ¼6.46) [r(28) ¼ 0.16; p ¼0.45] and with the RTs (mean ¼3111; SD ¼900) [r(28) ¼ 0.12; p ¼0.54] in the distance comparison task about the healthy group.
4. Discussion Our aim in the present work was to explore possible disturbances in obese people's mental body-part representations in two different tasks, based on tactile and visual sensory input. According to our results, obese people overestimated the size of the distances when tactilely perceived on the arm, and not when perceived on the abdomen more than did the healthy-weight group. About the abdomen, no difference in the tactilely based estimation emerged between the obese group and the healthyweight group. Thus, the observed pattern of overestimation was not a generalized effect but was body-part related; moreover, it was independent from gender, contrary to what was reported by Pearlson et al. (1981) and Cafri and Thompson (2004), according to which, particularly women tended to overestimate body size. The tactile estimation task allowed us to investigate the tactilely perceived body image (Keizer et al., 2011). The results of the present study suggest that this mental representation would be distorted in obesity. The estimation of a tactilely perceived distance on a body part's surface seems to be strictly correlated to the implicit size of the touched body part (Serino and Haggard, 2010; Longo et al., 2010). For example, since in our study the tactile stimulation was vertically applied on the arm, the overestimated distance on the arm would mirror a mental representation of the arm itself as longer than the real dimension, which is to say that obese people tend to enlarge the representation of their limb. Our results are in line with previous studies focused on the body-site adjustment procedure, according to which obese people are generally less accurate than healthy-weight people when asked to estimate the dimensions of body parts (Fisher, 1986; Cafri and Thompson, 2004) and they showed a general trend of
overestimation (Yamokoski, 1975; Garner et al., 1976; Kreitler and Chemerinsky, 1990). But why did obese people implicitly overestimate the arm's dimension and not that relative to the abdomen, even though they reported an equal rate of dissatisfaction and concern about slimming down for both of these body parts? From an anatomical point of view, the arm is more distal than the abdomen relative to the sagittal body plane, and our experience tells us how much it is involved in our interaction with objects in the peripersonal space (Rizzolatti et al, 1997), meaning that the portion of space around the body and its boundaries are deﬁned from the extension of the arm's action. Indeed the size of body parts, meaning the body's geometry, and the shapes of body parts are especially important and crucial when acting (De Vignemont, 2011; Graziano and Botvinick, 2002) since they deﬁne the arm's position and thus play a role in the act of reaching toward an object (Graziano and Botvinick, 2002). The overestimation of the arm's dimensions as a mental representation would reﬂect a general modiﬁcation of the body's geometry in order to take into account the real physical larger size of obese bodies, which is crucial during the generation of active movements in the environment towards external objects. This hypothesis reﬂects the multidimensional cognitive body representation (body matrix, Moseley et al., 2012), which integrates not only primary sensory input together with psychological and emotional components, but also the space around the body (Moseley et al., 2012) and the position of the objects around and with respect to the body (Ladavas et al., 1998; Holmes and Spence, 2004). Thus the dimensional representation of our bodies would affect our actions in the space, as recently reported by Keizer et al. (2013): the movement in space appeared to be based on body size information congruent with people's own perception of themselves instead of on the real body's dimensions (Keizer et al., 2013). Thus an interesting point for future research would be to explore deeply the role of the different representations of body parts in relation to the peripersonal space in cases of obesity by speciﬁc action-oriented tasks. The overestimation of the effector used for the action would affect the motor act; indeed it was reported that metric representation of the forearm is not substantially different when investigated through motor or perceptual tasks (Cardinali et al., 2011). In our study, the pattern of overestimation relative to the arm would implicitly refer to its larger cortical representation compared to the abdomen. In fact these body parts would be differentiated on the basis of neurophysiological variations and early processes mediated by primary somatosensory areas (Spitoni et al., 2010), such as the size of tactilely perceived receptive ﬁelds on the skin, the size of primary somatosensory representations, and the spatial quality of tactile sensations (Serino and Haggard, 2010). For example, the perceived distance between touches on a single skin surface is larger on regions of high tactile sensitivity than on those with lower acuity, according to the Weber's illusion effect (Weinstein, 1968; Taylor-Clarke et al., 2004; Longo and Haggard, 2011). However, Taylor-Clarke et al. (2004) hypothesized that an additional systematic process would correct for this early distortion. That is, this systematic process acts when the tactile information from the primary somatosensory representation is used in a secondary body representation and when size (as in the present task) and shape of the skin surface are computed. This representation refers to a pre-existing model of the actual metric properties of the body (Longo et al., 2010), derived from a combination of different somatosensory information (Longo and Haggard, 2012). Thus, primary and secondary body representations inﬂuence each other (Longo et al., 2010). The question is if the pattern of overestimation relative to the arm would be dependent on the characteristics of the primary somatosensory representation or of the distorted metric secondary body representation.
F. Scarpina et al. / Psychiatry Research 220 (2014) 960–969
Nevertheless, since an aberrant primary somatosensory perception was reported in eating disorders, such as in bulimia nervosa (Florin et al., 1988) and in anorexia nervosa (Keizer et al., 2012), and preliminary results about different pain sensitivity in obese patients (McKendall and Haier, 1983; Peltonen et al., 2003; Hitt et al., 2007; Somers et al., 2011), further investigations are required to investigate psychophysical assessments of primary somatosensory representation, such as tactile sensation and discrimination in cases of obesity. In the distance comparison task, participants were asked to visually imagine their own bodies and to evaluate the horizontal distances between different body parts (Smeets et al., 2009). These tasks were based on a visual body representation (Keizer et al., 2011). That is, subjects had to create a mental image of their body and to inspect it for its size. In the present study, the obese participants showed an inappropriate visual mental body image: they showed lower levels of accuracy than the healthy-weight participants did when they were asked to compare the extension of horizontal body distances and speciﬁcally, when the comparison involved body parts that were sensitive to preoccupation about size and shape (Molinari, 1995; Smeets et al., 2009). These results are in line with previous reports in the relevant literature, in which obese people were found to be generally less accurate than healthy-weight people when asked to estimate the dimension of body parts (Fisher, 1986; Cafri and Thompson, 2004); speciﬁcally we observed a high trend of overestimation (Yamokoski, 1975; Garner et al., 1976). In the present task, the participants (obese healthy-weight groups) needed more time to estimate the sensitive than the insensitive parts. This result mirrored what Smeets et al. (2009) hypothesized in their seminal work – that body parts that are sensitive to body shape concerns would take a relatively longer time to be scanned than body parts that are insensitive to such concerns (Smeets et al., 2009); however these results contrast the same authors' ﬁndings among their sample, according to which, distance-difference evaluations involving sensitive parts required less time. Smeets et al. (2009) suggested that people who have signiﬁcant shape concerns about speciﬁc body parts inspected them more often than other parts, constructing a more accessible mental representation (as reﬂected by lower reaction times). Another possible explanation about the increasing of reaction times for sensitive parts with respect to the insensitive body parts would be related to the topographical linear relation between their real physical distance and the imagined representation (Kosslyn, 1980; Smeets et al., 2009). More time is required to inspect a larger distance than a shorter one between objects in the mental representation, just as in the real world. Considering the pairs presented in the distance comparison task, one would notice that generally the horizontal distances between waists, hips, and thighs are larger than those between ears and knees, which, on the other hand, would not be true about shoulders, armpits, and elbows. This difference would partially explain why sensitive parts (i.e. longer distances) required more time than insensitive parts (i.e. shorter distances) to be estimated for both the obese and healthy-weight groups. In this study, in obese group the tactilely based overestimation of the distance was related to a lower amount of time to mentally scan the horizontal distances. In their study about body image representation in anorexia nervosa, Keizer et al. (2011) reported that the error in the tactile size estimation task increased while accuracy on the distance comparison task decreased, which suggests a relationship between the tactilely based and the visually based body-part representations. A similar relationship was also found in Kreitler and Chemerinski's (1990) study, in which subjects showed the size of waist, the hips, the mouth, and the face by means of their ﬁngers or hands, while they stood at their ease with closed eyes. According to the reported results, the size overestimation of the examined body parts was related to the
larger dimensions of the same parts when obese people were required to draw human ﬁgures (the human ﬁgure drawing test). This pattern of results was interpreted as being suggestive of a lower developmental level of the body image in obese individuals, without differentiation between the kinaesthetic and the graphic representations (Kreitler and Chemerinski, 1990). According to the dyadic model about how the body is mentally represented (Head and Holmes, 1911; Paillard, 1999; Gallagher, 2005), the pattern of results described by Kreitler and Chemerinski (1990), as our results would also, are suggestive of a general distortion in mental body representation in obesity. This distortion would be tracked down in multiple modalities, since mental body representation is constructed from and reciprocally inﬂuenced by input sources (Serino and Haggard, 2010). As suggested by Keizer et al. (2011), relative to anorexia nervosa patients, the body representation disturbance in the tactile and visual modalities would result from top-down inﬂuences of body dissatisfaction on the mental body representation in case of obesity as well. Different from what we have found in the obese population, the tendency to overestimate distance based on tactile stimuli was reported to be generalized among anorexia nervosa patients (Keizer et al., 2011, 2012). The different pattern of results in the estimation of body parts and consequently the characteristics of the mental body representation would mirror the differences in emotional conﬂicts between anorexia nervosa and obesity, suggesting that the direct comparison between these two conditions would not be pertinent. When people were asked to respond about speciﬁc parts of their body, their feelings and cognitive concepts relative to those body parts (Shontz, 1969) and preoccupations about size and shape (Molinari, 1995; Smeets et al., 2009) were enhanced. This hypothesis would apply not only to pathological populations, but also for healthy-weight people. This was sketched out in the rating concerns about the arm and the abdomen expressed by the participants in the present study. The obese participants expressed a high level of preoccupation about both body parts; moreover, the healthy-weight participants declared to be more concerned that their abdomen would not slim down than they were about their arm not slimming down, which ﬁndings were similar to those reported after similar questions by Keizer et al. (2012). Moreover, the ﬁndings mirrored both the perceptual judgment relative to the abdomen reported by two groups in the tactile task and the similar increase of amount on time that participants needed to judge the horizontal extension of sensitive body parts in the distance comparison task. In Western societies, slenderness is generally associated with positive characteristics like happiness, success, youthfulness, and social acceptability, while being overweight is linked to laziness, lack of willpower, and being out of control: being overweight (for both men and women) is seen as physically unattractive (Grogan, 2008); nonconformity to ideal slim bodies has a variety of negative social consequences (Grogan, 2008). Thus, considering the previous issues, obese people and healthy-weight people would share an amount of concern about speciﬁc body parts, like the abdomen, for different reasons: the ﬁrst group is worried about not achieving the goal of being slim and the second one is worried about diverging from the social imperative of having ideal body dimensions (Grogan, 2008). In conclusion, a distortion in the size estimation of body parts emerged in the visually and tactilely based mental body representations among obese people; however, their overestimation is not a generalized tendency but was body-part related. Since it has been reported as a residual body image misperception following weight loss (“phantom fat” phenomenon, Cash, 1990), suggesting that the mental body representation is not strictly related to the physical dimension but is intimately related to self-perception; additionally the mental body representation is possibly part of the relationship between perceptual misjudgement and compliance
F. Scarpina et al. / Psychiatry Research 220 (2014) 960–969
with the rehabilitation program for losing weight (Sarwer et al., 1998; Sarwer and Thompson, 2002; Foster et al., 1997; Ramirez and Rosen, 2001; Riva et al., 2006; Villa et al., 2009; Castelnuovo et al., 2010; Cernelic-Bizjak and Jenko-Praznikar, 2014) and with quality of life (Lo Coco et al., 2014).
Acknowledgment We thank Annalisa Carrera and Stefania Corti for helping in the recruitment of the clinical population; Gabriella Meloni for her assistance in data collection of the healthy-weight participants. References Bell, C., Kirkpatrick, S.W., Rinn, R.C., 1986. Body image of anorexic, obese, and normal females. Journal of Clinical Psychology 42, 431–439. Cafri, G., Thompson, J.K., 2004. Measuring male body image: a review of the current methodology. Psychology of Men & Masculinity 5 (1), 18–29. Cardinali, L., Brozzoli, C., Urquizar, C., Salemme, R., Roy, A.C., Farnè, A., 2011. When action is not enough: tool-use reveals tactile-dependent access to body schema. Neuropsychologia 49, 3750–3757. Cash, T.F., 1990. The psychology of physical appearance: aesthetics, attributes, and images. In: Cash, T.F., Prunzinsky, T. (Eds.), Body Images: Development, Deviance, and Change. The Guilford Press, New York, NY, pp. 51–79. Castelnuovo, G., Manzoni, G.M., Cuzziol, P., Cesa, G.L., Tuzzi, C., Villa, V., Liuzzi, A., Petroni, M.L., Molinari, E., 2010. TECNOB: study design of a randomized controlled trial of a multidisciplinary telecare intervention for obese patients with type-2 diabetes. BMC Public Health 10, 204. Cernelic-Bizjak, M., Jenko-Praznikar, Z., 2014. Impact of negative cognitions about body image on inﬂammatory status in relation to health. Psychology & Health 29 (3), 264–278. Collins, J.K., Beaumont, P.J.V., Touyz, S.W., Krass, J., Thompson, P., Philips, T., 1987. Variability in body shape perception in anorexic, bulimic, obese, and control subjects. International Journal of Eating Disorders 6, 633–638. De Vignemont, F., 2010. Body schema and body image – Pros and cons. Neuropsychologia 48, 669–680. De Vignemont, F., 2011. Embodiment, ownership and disownership. Consciousness and Cognition 20 (1), 82–93. De Vignemont, F., Ehrsson, H.H., Haggard, P., 2005. Bodily illusions modulate tactile perception. Current Biology 15, 1286–1290. De Vignemont, F., Farnè, A., 2010. Incorporer des objets et des membres factices: Quelle différence? Revue de Neuropsychologie 2, 203–211. Denis, M., 2008. Assessing the symbolic distance effect in mental images constructed from verbal description: a study on individual differences in the mental comparison of distance. Acta Psychologica 127, 197–210. Dijkerman, H.C., De Haan, E.H.F., 2007. Somatosensory processes subserving perception and action. Behavioral and Brain Sciences 30, 189–239. Docteur, A., Urdapilleta, I., Defrance, C., Raison, J., 2010. Body perception and satisfaction in obese, severely obese, and healthy-weight female patients. Obesity 18, 1464–1465. Fisher, S., 1986. Development and Structure of the Body Image. vols. 1 and 2. Lawrence Erlbaum Associates, Hillsdale, NJ. Florin, I., Franzen, U., Meier, M., Schneider, S., 1988. Pressure sensitivity in bulimic women: a contribution to research in body image distortion. Journal of Psychosomatic Research 32 (4-5), 439–444. Folstein, M.F., Folstein, S.E., McHugh, P.R., 1975. “Mini-mental State.” A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research 12, 189–198. Foster, G.D., Wadden, T.A., Vogt, R.A., 1997. Body image in obese women before, during, and after weight loss treatment. Health Psychology 16, 226–229. Friedman, M.A., Brownell, K.D., 1995. Psychological correlates of obesity: moving to the next research generation. Psychological Bulletin 117, 3–20. Friedman, K.E., Reichmann, S.K., Costanzo, P.R., Musante, G.J., 2002. Body image partially mediates the relationship between obesity and psychological distress. Obesity Research 10 (1), 33–41. Gallagher, S., 2005. How the Body Shapes the Mind. Oxford University Press, Oxford, New York. Gardner, R.M., Martinez, R., Sandoval, Y., 1987. Obesity and body image: an evaluation of sensory and non-sensory components. Psychological Medicine 17, 927–932. Gardner, R.M., Gallegos, V., Martinez, R., Espinoza, T., 1989. Mirror feedback and judgments of body size. Journal of Psychosomatic Research 33, 603–607. Garner, D.M., Garﬁnkel, P.E., Stancer, H.C., Moldofsky, H., 1976. Body image disturbance in anorexia nervosa and obesity. Psychosomatic Medicine 38, 327–336. Glucksman, M.L., Hirsch, J., 1969. The response of obese patients to weight reduction III. The perception of body size. Psychosomatic Medicine 31 (1), 1–7. Graziano, M.S.A., Botvinick, M.M., 2002. How the brain represents the body: Insight neurophysiology and psychology. In: Prinz, W., Hommel, B. (Eds.), Common
Mechanisms in Perception and Action: Attention and Performance XIX. Oxford University Press, Oxford, pp. 136–157. Grilo, C.M., White, M.A., Masheb, R.M., 2012. Signiﬁcance of overvaluation of shape and weight in an ethnically diverse sample of obese patients with binge-eating disorder in primary care settings. Behaviour Research and Therapy 50 (5), 298–303. Grogan, S., 2008. Body Image. Understanding body dissatisfaction in men, women, and children, 2nd ed. Taylor & Francis e-Library, New York, NY. Head, H., Holmes, H.G., 1911. Sensory disturbances from cerebral lesions. Brain 34, 102–254. Hitt, H.C., McMillen, R., Thortnton-Neaves, T., Koch, K., Cosby, A.G., 2007. Comorbidity of obesity and pain in a general population: results from the southern pain prevalence study. The Journal of Pain 8 (5), 430–436. Holmes, N., Spence, C., 2004. The body schema and multisensory representation (s) of peripersonal space. Cognitive Processing 5 (2), 94–105. Kalliopuska, M., 1982. Body-image disturbances in patients with anorexia nervosa. Psychological Reports 51, 715–722. Keizer, A., Smeets, M.A.M., Dijkerman, H.C., van den Hout, M., Klugkist, I., van Elburg, A., Postma, A., 2011. Tactile body image disturbance in anorexia nervosa. Psychiatry Research 190, 115–120. Keizer, A., Smeets, M.A.M., Dijkerman, H.C., van Elburg, A., Postma, A., 2012. Aberrant somatosensory perception in anorexia nervosa. Psychiatric Research 200 (2–3), 530–537. Keizer, A., Smeets, M.A.M., Dijkerman, H.C., Uzunbajakau, S.A., van Elburg, A., Postma, A., 2013. Too fat to ﬁt through the door: ﬁrst evidence for disturbed body-scaled action in anorexia nervosa during locomotion. PLOS One 8 (5), e64602. Kosslyn, S.M., 1980. Image and Mind. Harvard University Press, Cambridge. Kreitler, S., Chemerinsky, A., 1990. Body-image disturbance in obesity. International Journal of Eating Disorders 9 (4), 409–418. Ladavas, E., Zeloni, G., Farnè, A., 1998. Visual peripersonal space centred on the face in humans. Brain 121, 2317–2326. Legenbauer, T., Vocks, S., Betz, S., Baguena Puigcerver, M.J., Benecke, A., Troje, N.F., Ruddel, H., 2011. Differences in the nature of body image disturbances between female obese individuals with versus without a comorbid binge eating disorder: an exploratory study including static and dynamic aspects of body image. Behavior Modiﬁcation 35 (2), 162–186. Lo Coco, G., Salerno, L., Bruno, V., Caltabiano, M.L., Ricciardelli, L.A., 2014. Binge eating partially mediates the relationship between body image dissatisfaction and psychological distress in obese treatment seeking individuals. Eating Behaviors 15, 45–48. Longo, M.R., Azaňòn, E., Haggard, P., 2010. More than skin deep: body representation beyond primary somatosensory cortex. Neuropsychologia 48, 655–668. Longo, M.R., Haggard, P., 2011. Weber's illusion and body shape: anisotropy of tactile size perception on the hand. Journal of Experimental Psychology: Human Perception and Performance 37 (3), 720–726. Longo, M.R., Haggard, P., 2012. Implicit body representations and the conscious body image. Acta Psychologica 141 (2), 164–168. Mathôt, S., Schreij, D., Theeuwes, J., 2012. Open sesame: an open-source, graphical experiment builder for the social sciences. Behavior Research Methods 44, 314–324. McElroy, S.L., Kotwal, R., Malhotra, S., Nelson, E.B., Keck Jr., P.E., Nemeroff, C.B., 2004. Are mood disorders and obesity related? A review for the mental health professional. Journal of Clinical Psychiatry 65, 634–651. McKendall, M.J., Haier, R.J., 1983. Pain sensitivity and obesity. Psychiatric Research 8 (2), 119–125. Molinari, E., 1995. Body-size estimation in anorexia nervosa. Perceptual and Motor Skills 81, 23–31. Morgan, J.F., Reid, F., Lacey, J.H., 1999. The SCOFF questionnaire: assessment of a new screening tool for eating disorders. British Medical Journal 4 319 (7223), 1467–1468. Moseley, G.L., Gallace, A., Spence, C., 2012. Bodily illusions in health and disease: Physiological and clinical perspectives and the concept of a cortical ‘body matrix’. Neuroscience & Biobehavioral Reviews 36 (1), 34–46. Nicoli, M.G., Junior, R.D., 2011. Binge eating disorder and body image perception among university students. Eating Behaviors 12 (4), 284–288. Paillard, J., 1999. Body schema and body image: a double dissociation in deafferented patients. In: Gautchev, G.N., Mori, G.N., Massion, S.J. (Eds.), Motor Control, Today and Tomorrow. Academic Publishing House, Soﬁa, pp. 197–214. Pannocchia, L., Fiorino, M., Giannini, M., Vanderlinden, J., 2011. A psychometric exploration of an Italian translation of the SCOFF questionnaire. European Eating Disorders Review: The Journal of the Eating Disorders Association 19, 371–373. Pearlson, G.D., Flournoy, L.H., Simonson, M., Slavney, P.R., 1981. Body image in obese adults. Psychological Medicine 11 (1), 147–154. Peltonen, M., Lindroos, A.K., Torgenson, J.S., 2003. Musculoskeletal pain in the obese: a comparison with a general population and long-term changes after conventional and surgical obesity treatment. Pain 104 (3), 549–557. Probst, M., Van Coppenolle, H., Vandereycken, W., Goris, M., 1992. Body image assessment in anorexia nervosa patients and university students by means of video distortion: a reliability study. Journal of Psychosomatic Research 36 (1), 89–97. Ramirez, E.M., Rosen, J.C., 2001. A comparison of weight control and weight control plus body image therapy for obese men and women. Journal of Consulting and Clinical Psychology 69, 440–446.
F. Scarpina et al. / Psychiatry Research 220 (2014) 960–969
Rizzolatti, G., Fadiga, L., Fogassi, L., Gallese, V., 1997. The space around us. Science 277 (5323), 190–191. Riva, G., Bacchetta, M., Cesa, G., Conti, S., Castelnuovo, G., Mantovani, F., Molinari, E., 2006. Is severe obesity a form of addiction? Rationale, clinical approach, and controlled clinical trial. Cyberpsychology & Behavior: The Impact of the Internet, Multimedia and Virtual Reality on Behavior and Society 9, 457–479. Sarwer, D.B., Wadden, T.A., Foster, G.D., 1998. Assessment of body image dissatisfaction in obese women: speciﬁcity, severity, and clinical signiﬁcance. Journal of Consulting and Clinical Psychology 66 (4), 651–654. Sarwer, D.B., Thompson, J.K., 2002. Obesity and body image disturbance. In: Wadden, T.A., Stunkard, A.J. (Eds.), Handbook of Obesity Treatment. Guilford Press, New York, NY, pp. 447–464. Schwartz, M.B., Brownell, K.D., 2004. Obesity and body image. Body Image 1 (1), 43–56. Sedda, A., Scarpina, F., 2012. Dorsal and ventral streams across sensory modalities. Neuroscience Bulletin 28 (3), 291–300. Serino, A., Haggard, P., 2010. Touch and the body. Neuroscience and Biobehavioral Reviews 34, 224–236. Shontz, F.C., 1969. Perceptual and Cognitive Aspects of Body Experience. Macmillan, New York. Slade, P.D., Russell, G.F.M., 1973. Awareness of body dimensions in anorexia nervosa: cross sectional and longitudinal studies. Psychological Medicine 3, 188–199. Smeets, M.A.M., Klugkist, I., van Rooden, S., Anema, H., Postma, A., 2009. Mental body distance comparison: a tool for assessing clinical disturbances in visual body image. Acta Psychologica 132, 157–165.
Somers, T.J., Wren, A.A., Keefe, F.J., 2011. Understanding chronic pain in older adults: abdominal fat is where it is at. Pain 152 (1), 8–9. Spitoni, G.F., Galati, G., Antonucci, G., Haggard, P., Pizzamiglio, L., 2010. Two forms of touch perception in the human brain. Experimental Brain Research 207 (3–4), 185–195. Taylor-Clarke, M., Jacobsen, P., Haggard, P., 2004. Keeping the word a constant size: object constancy in human touch. Nature Neuroscience 7, 219–220. Tiggeman, M., Rothblum, E.D., 1988. Gender differences in social consequences of perceived overweight in the United States and Australia. Sex Roles 18, 75–86. Valtolina, G., 1998. Body-size estimation by obese subjects. Perceptual and Motor Skills 86, 1363–1374. Villa, V., Manzoni, G.M., Pagnini, F., Castelnuovo, G., Cesa, G.L., Molinari, E., 2009. Do coping strategies discriminate eating disordered individuals better than eating disorder features? An explorative study on female inpatients with anorexia and bulimia nervosa. Journal of Clinical Psychology in Medical Settings 16, 297–303. Weinstein, S., 1968. Intensive and extensive aspects of tactile sensitivity as a function of body part, sex, and laterality. In: Kenshalo, D.R. (Ed.), The Skin Senses. Charles C. Thomas, Springﬁeld, IL, pp. 195–222. Wingate, B.A., Christie, M.K., 1978. Ego strength and body image in anorexia nervosa. Journal of Psychomatic Research 22, 201–204. World Medical Association, 1991. Declaration of Helsinki. British Medical Journal 302, 1194. Yamokoski, T.B., 1975. Distractibility and body image in juvenile-onset and adultonset obese individuals. Dissertation Abstracts International 36, 4715–4716.