Short-term fasting selectively influences impulsivity in healthy individuals

Maxine Howard, Jonathan P. Roiser, Sam J. Gilbert, Paul W. Burgess, Peter Dayan, Lucy SerpellbioRxiv 2018 · DOI

Abstract

Previous research has shown that short-term fasting in healthy individuals (HIs) is associated with changes in risky decision-making. The current experiment was designed to examine the influence of short-term fasting in HIs on four types of impulsivity: reflection impulsivity, risky decision-making, delay aversion, and action inhibition. HIs were tested twice, once when fasted for 20 hours, and once when satiated. Participants demonstrated impaired action inhibition when fasted, committing significantly more errors of commission during a food-related Affective Shifting Task. Participants also displayed decreased reflection impulsivity when fasted, opening significantly more boxes during the Information Sampling Task (IST). There were no significant differences in performance between fasted and satiated sessions for risky decision-making or delay aversion. These findings may have implications for understanding eating disorders such as Bulimia Nervosa (BN). Since individuals with BN undergo periods of short-term fasting, inconsistent findings in BN research could be due to differences in the levels of satiation of participants.

Introduction: State vs. Trait Impulsivity

Impulsivity is not a monolithic trait but a "multifactorial construct"—a collection of distinct cognitive processes. This distinction is vital for neuropsychological research, particularly regarding eating disorders like Bulimia Nervosa (BN). While BN is clinically characterized by poor impulse control, laboratory studies comparing BN patients to healthy controls have yielded inconsistent results.

The authors of this study hypothesized that these inconsistencies might stem from a confusion between "state" and "trait" variables. Because individuals with BN frequently engage in restrictive fasting, their performance on cognitive tests may reflect a temporary physiological state rather than an underlying personality trait. By observing how a 20-hour fast affects healthy individuals, this study aimed to isolate how acute hunger alters specific dimensions of impulsivity.

Methodology

The study utilized a within-subjects, repeated-measures design involving 33 healthy female participants. Each participant was tested under two conditions: once after a 20-hour fast and once in a satiated state. To validate the fasting state, researchers measured blood glucose levels and utilized self-reported hunger scales.

Participants performed four behavioral tasks targeting specific impulsivity subtypes:

  1. Information Sampling Task (IST): Measures reflection impulsivity (the amount of information gathered before making a decision).
  2. Affective Shifting Task (AST): Measures action inhibition (the ability to stop a motor response), using food and household imagery.
  3. Choice x Risk Task (CRT): Measures risky decision-making (preference for high-risk/high-reward vs. low-risk/low-reward).
  4. Temporal Discounting Task (TDT): Measures delay aversion (the tendency to prefer smaller immediate rewards over larger delayed rewards).

Key Results

Physiological Impact

The 20-hour fast successfully induced a significant metabolic shift, with blood glucose levels dropping to an average of 4.06 mmol/L compared to 4.90 mmol/L in the satiated state.

Selective Impairment of Action Inhibition

Fasting significantly impaired "bottom-up" motor control. In the Affective Shifting Task, fasted participants made significantly more "errors of commission" (failing to inhibit a response) compared to when they were full ($M=1.55$ vs. $M=1.19$). Interestingly, this struggle was most pronounced when participants were presented with food-related images during "shift" blocks, suggesting that hunger compromises the ability to inhibit responses to food cues when cognitive demand is high.

The "Information Trap" in Reflection Impulsivity

Contrary to the common assumption that hunger leads to rashness, fasting actually decreased reflection impulsivity. In the "Fixed Win" condition of the IST, fasted participants opened significantly more boxes ($M=17.07$) than they did when satiated ($M=13.73$). Rather than rushing to a conclusion, hungry participants became more cautious, gathering extra data before committing to a choice.

Stability in Economic Decision-Making

Fasting had no significant effect on "top-down" economic choices. There were no measurable differences in risky decision-making (CRT) or the rate at which participants discounted future monetary rewards (TDT) between the fasted and satiated sessions.

Discussion and Clinical Implications

These findings suggest that some "impulsive" behaviors observed in clinical populations may be the biological signature of a fasted brain rather than a fundamental character deficit. The discovery that fasting selectively influences the IST—a measure frequently used in BN research—provides a potential explanation for previous inconsistent findings: the results may have been confounded by the participants' immediate hunger levels.

However, the unexpected finding that fasted individuals gather more information (decreased reflection impulsivity) requires further investigation. This could be attributed to "set-shifting" difficulties—where participants get stuck in the repetitive habit of opening boxes—or a loss of "central coherence," where the brain focuses on minute details at the expense of the larger picture.

Limitations and Future Directions

  • Gender Generalizability: The study was limited to female participants; metabolic stress may affect men differently.
  • Reward Modality: The risk and delay tasks utilized monetary rewards. While fasting may not change the value of money, it likely alters the valuation of food rewards (e.g., immediate vs. delayed caloric intake).
  • Causal Mechanisms: It remains to be determined whether these cognitive shifts are driven specifically by low blood glucose or by the broader psychological experience of hunger and fatigue.

Summary updated

Abstract

Abstract Previous research has shown that short-term fasting in healthy individuals (HIs) is associated with changes in risky decision-making. The current experiment was designed to examine the influence of short-term fasting in HIs on four types of impulsivity: reflection impulsivity, risky decision-making, delay aversion, and action inhibition. HIs were tested twice, once when fasted for 20 hours, and once when satiated. Participants demonstrated impaired action inhibition when fasted; committing significantly more errors of commission during a food-related Affective Shifting Task. Participants also displayed decreased reflection impulsivity when fasted, opening significantly more boxes during the Information Sampling Task (IST). There were no significant differences in performance between fasted and satiated sessions for risky decision-making or delay aversion. These findings may have implications for understanding eating disorders such as Bulimia Nervosa (BN). Although BN has been characterised as a disorder of poor impulse control, inconsistent findings when comparing individuals with BN and HIs on behavioural measures of impulsivity question this characterisation. Since individuals with BN undergo periods of short-term fasting, the inconsistent findings could be due to differences in the levels of satiation of participants. The current results indicate that fasting can selectively influence performance on the IST, a measure of impulsivity previously studied in BN. However, the results from the IST were contrary to the original hypothesis and should be replicated before specific conclusions can be made.
BibTeX
@misc{howard2018short,
  author = {Maxine Howard and Jonathan P. Roiser and Sam J. Gilbert and Paul W. Burgess and Peter Dayan and Lucy Serpell},
  title = {{Short-term fasting selectively influences impulsivity in healthy individuals}},
  year = {2018},
  doi = {10.1101/468751},
  howpublished = {bioRxiv},
}
Figure