Diagnostic validity of behavioural and psychometric impulsivity measures: An assessment in adolescent and adult populations
Introduction
Impulsivity denotes a tendency to act without considering consequences, where responses are rapid and without thought, and has been identified to develop with age (Steinberg, 2004), thought to peak at adolescence and decreasing thereafter with the development of cognitive functioning associated with the maturation of the prefrontal cortex (Galvan et al., 2006). Teenage impulsiveness stems from an imbalance between two interconnected brain networks: the incentive processing system, associated with the process of rewards and punishment, and the cognitive control system, which is associated with logical reasoning and impulse regulation (Steinberg, 2008). The incentive processing system begins to develop rapidly with the onset of puberty causing teenagers to overweight rewards, leading to sensation-seeking behaviour (Ernst et al., 2005). In contrast the cognitive control system has a much slower developmental rate, which continues well into the twenties (Steinberg, 2004). The mid-adolescence period in particular is a particularly vulnerable period when the disparity between the two systems is largest.
Measuring impulsivity has important consequences with the healthcare field, where recent evidence in neuropsychology and consumer behaviour has connected impulsivity and addiction behaviour with obesity (Balodis et al., 2014). Obese individuals demonstrate less impulsive control than the healthy population (Nederkoorn, 2014), where obesity is a direct outcome of the neurological and psychological factors that negatively influence an individual's rational decision-making in healthy dieting and exercising decisions (Takahashi, 2004, Weller et al., 2008). Such empirical evidence therefore suggests that measuring impulsivity has clinical advantages in identifying at risk impulsive individuals who are likely to become clinically obese.
There have been numerous investigations into the relationship of obesity and impulsivity. For instance, Sutin, Ferrucci, Zonderman, and Terracciano (2011) identified a relationship between personality measures and impulsivity, specifically measures relating to impulsive behaviour. The researchers found that impulsivity was a major factor in obesity, with an 11 kg weight disparity between the highest and lowest 10% scores on impulsivity measures.
A study to investigate the direct link between impulsivity and obesity was conducted by Guerrieri, Nederkoorn, Schrooten, Martijn, and Jansen (2009), who manipulated impulsive behaviour in participants using a priming task, the researches found a heightened food intake compared to those with higher levels of impulsivity compared to those who were not primed.
These findings demonstrate how important accurate measures of impulsivity are within a clinical setting. However previous research fail to adopt overarching measures of impulsivity across all ages. For instance, Nederkoorn, Smulders, Havermans, Roefs, and Jansen (2006) examine impulsivity through psychometric (Barrat Impulsivity Scale [BIS] and the Hyperactivity/Impulsivity subscale) in addition to behavioural measures (Stop Signal Task and the Door Opening Task), the study uses a cross-sectional design but focuses on younger population. The researchers found significantly more impulsive behaviours in obese children compared to healthy in the Stop Signal Task, Door Opening Task and the Fun Seeking subscale of the BIS. However such findings are only generalizable to a young population and do not extend towards adults. A secondary aspect is that the use of a cross-sectional design only ascertains an association of obesity and impulsivity, but does not pertain to how impulsive behaviour fairs over time.
A second example is denoted by Rasmussen, Lawyer, and Reilly (2010) examining discounting behaviour of impulsivity in overweight participants. In this study, participants were asked which of two options they prefer, a smaller-certain reward or a larger but uncertain reward. The authors found that individuals with a higher percentage of body fat would discount hypothetical food choices, opting for smaller-sooner bites against larger-later, despite not controlling for hunger, suggesting that impulsive factors could be a cause of obesity. Such findings are consistent with other evidence on discounting behaviour (Davis, Levitan, Muglia, Bewell, & Kennedy, 2004). However a major shortcoming of this study is the use of a sole measure (discounting behaviour) in an adult population. As such, the results are difficult to integrate with other measures (e.g. Stop Signal Task) and generalize towards other younger populations.
A third example is from Houben, Nederkoorn, and Jansen (2014), the authors used the CANTAB Stop Signal Task (SSRT) (Logan, Schachar, & Tannock, 1997), a measure of inhibitory control and a variation which they devised using food-related stimuli. Houben and colleagues found that obese adults (mean age: 26.17 years) demonstrated more impulsive behaviour for food-related response compared to general cues, suggesting that obesity could be characterized by a lack of impulse control to food cues. The researchers here have also utilized a single measure of impulsivity, although they do investigate general and food-related impulsivity. However it is difficult to ascertain how such findings correspond to discounting behaviour and self-reported impulsivity.
Longitudinal studies of impulsive measures have been investigated, for example Kulendran et al. (2014) compared obese and healthy matched adolescents, through a pretest/posttest design using behavioural and psychometric measures. The behavioural measures denoted the Stop Signal Task and a Temporal Discounting Task. The researchers also utilized psychometric, personality measures to assess individual characteristics that predispose impulsive behaviour, this was achieved with the Junior Temperament and Character Inventory (Cloninger, Przybeck, Svrakic, & Wetzel, 1994). The use of psychometric scales offers a novel touch to the design, by expanding the impulsivity measurement framework to include behavioural, neurological and psychological domains. The researchers identified higher discount rates, as well as decreased inhibitory control, in obese participants compared to healthy individuals. Obese adolescents who attended weight loss camp experiences a significant reduction in Body Mass Index (BMI) and impulsivity measures, where age was shown to significantly moderate such effects. However, one major absence of this study is the limited scope of the sample based on adolescent population. Such findings may not apply and generalize to an older population where the cognitive control system is fully developed and integration of the control and incentive based systems are established.
To date, no one study has provided a comprehensive validation of both behavioural and psychological measures of impulsivity amongst obese individuals across both adult and adolescent years. Several studies utilize some measures of impulsivity amongst obese individuals (Nederkoorn, 2014), but these studies often involve either behavioural and psychometric but within a limited sample, or across samples but with limited measures (strictly behavioural or psychometric). No single study uses behavioural measures (discounting and inhibitory control measures), in addition to psychometric based measures across both adolescents and adults.
Assessing the measurement of impulsivity in obese participants actively seeking weight management against a normal, healthy weight subgroup will help to validate the use of the chosen measurement tools of impulsivity and delineate the construct further.
In order to study this, we designed a cross-sectional study using adult and adolescent based populations with varying levels of obesity, and incorporated behavioural and psychometric based measures to identify an overarching, impulsive measurement framework.
Section snippets
Sample
The sample collected was split across five groups, initially across age (adolescent and adult participants), then divided again into varying degrees of weight. Adolescents comprised of an obese sample and a healthy sample matched by age and sex were recruited from a local secondary school, whilst the adult sample was divided into an obese bariatric sample, obese individuals who required lifestyle management as a weight-reduction interventions, and healthy weight adults. Adults were not matched
Assumptions
Outliers were winsorized, being replaced with 95th percentile values (Tukey & McLaughlin, 1963). Assumptions of ANOVA were examined, where Temporal Discounting, D (202) = 0.126, p < 0.001; SSRT scores, D (202) = 0.094, p < 0.001; and BMI scores were non-normally distributed, D (201) = 0.183, p < 0.001. As such the variables were transformed by using a two-step transformation technique put forward by Templeton (2011). The newly transformed data identified normality in SSRT scores D (200) = 0.028, p = 0.200; and
Discussion
This paper addressed the development and validation of impulsive measures amongst obese individuals, amongst adolescent and adult clinical populations. The SSRT of the Stop Signal Task sufficiently differentiated between the five groups tested. It was also the only behavioural measure to correlate significantly with BMI. Although there was a difference in impulsivity as expected between the adult groups, there was no significant difference between individuals undergoing lifestyle management and
Conflict of interest
The authors declare no conflict of interest.
6. Acknowledgements
No funding source was used for this project.
References (28)
- et al.
Amygdala and nucleus accumbens in responses to receipt and omission of gains in adults and adolescents
NeuroImage
(2005) - et al.
Reliability and validity of the Italian version of the temperament and character inventory—Revised in an outpatient sample
Comprehensive Psychiatry
(2007) - et al.
Inducing impulsivity leads high and low restrained eaters into overeating, whereas current dieters stick to their diet
Appetite
(2009) - et al.
The reliability and validity of the junior temperament and character inventory
Comprehensive Psychiatry
(2004) - et al.
Impulsivity and test meal intake in obese binge eating women
Appetite
(2004) - et al.
Impulsivity in obese women
Appetite
(2006) - et al.
Percent body fat is related to delay and probability discounting for food in humans
Behavioural Processes
(2010) A social neuroscience perspective on adolescent risk-taking
Developmental Review
(2008)- et al.
Obese women show greater delay discounting than healthy-weight women
Appetite
(2008) - et al.
A pilot study linking reduced fronto-striatal recruitment during reward processing to persistent bingeing following treatment for binge-eating disorder
International Journal of Eating Disorders
(2014)
Gender differences in cognitive vulnerability to depression and behavior problems in adolescents
Journal of Abnormal Child Psychology
The adolescent brain
Annals of the New York Academy of Sciences
The Temperament and Character Inventory (TCI): A guide to its development and use
Decision-making deficits and overeating: A risk model for obesity
Obesity Research
Cited by (13)
Temporal discounting as a candidate behavioral marker of obesity
2021, Neuroscience and Biobehavioral ReviewsCitation Excerpt :More striking differences were noticed in the $50,000 hypothetical task, where for healthy weight individuals, a 50 % loss of value was seen at 36 months, while for obese individuals, this same percentage loss was seen at 3 months. The ability of TD to distinguish individuals with obesity from healthy weight individuals has been replicated in six subsequent studies of adolescents (Feda et al., 2015; Fields et al., 2013, 2011; Kulendran et al., 2016, 2014; Verdejo-García et al., 2010) and 19 studies of adults (Bickel et al., 2014a; Buono et al., 2015; Davis et al., 2010; DeHart et al., 2020; Garza et al., 2016; Graham Thomas et al., 2015; Hendrickson and Rasmussen, 2017; Jarmolowicz et al., 2014; Lawyer et al., 2015; Mole et al., 2015; Morys et al., 2018; Myers et al., 2020; Price et al., 2016; Simmank et al., 2015; Soofi et al., 2019; Steward et al., 2017; Stoklosa et al., 2018; Syan et al., 2019; Zimmerman et al., 2018). One study in adult undergraduate students showed that obese individuals had significantly greater TD rates when compared to underweight individuals, but not to healthy weight individuals (Wainwright et al., 2018).
Anatomy and disorders of frontal lobe functions: Higher-order functions
2021, Encyclopedia of Behavioral Neuroscience: Second EditionA review of effort-based decision-making in eating and weight disorders
2021, Progress in Neuro-Psychopharmacology and Biological PsychiatryCitation Excerpt :Eight studies demonstrate no difference in delay discounting in ED/OB/OW and healthy controls (see Table 2). When comparing across ages, older participants show less impulsive choices than young adults, preferring delayed rewards over immediate ones (Buono et al., 2015; Kulendran et al., 2016). Whereas the majority of studies use monetary rewards (N = 27), 3 studies utilized food, sedentary activities and discount vouchers as alternative reward stimuli (Manwaring et al., 2011; Schiff et al., 2016; Hendrickson and Rasmussen, 2013).
Changes in visual attention towards food cues after obesity surgery: An eye-tracking study
2020, Journal of Psychiatric ResearchCitation Excerpt :Studies using functional magnetic resonance imaging (fMRI) revealed that individuals with obesity compared to those with normal weight show both food-specific hyper-activation in brain areas linked to reward processing (orbitofrontal cortex, OFC) and visual attention (posterior cingulate cortex, inferior parietal lobe), as well as hypo-activation of areas related to inhibitory control (prefrontal cortex, PFC), resulting in lower dietary self-regulation and increased risk for overeating (Devoto et al., 2018; Lowe et al., 2019; Stice and Burger, 2019). Notably, experimental evidence indicated greater impulsivity the higher the BMI with pre-bariatric adults showing lower inhibitory control than patients with obesity undergoing behavioral weight-loss treatment (Kulendran et al., 2016). At the same time, the odds for recurrent binge eating, characterized by experiencing loss of control over eating, are 13 times greater in those with obesity class III (BMI ≥ 40 kg/m2) relative to those with obesity class I (BMI = 30–34.9 kg/m2; Duncan et al., 2017).
Executive function performance in obesity and overweight individuals: A meta-analysis and review
2018, Neuroscience and Biobehavioral Reviews