A recent study published in the Journal of Behavioral Addictions challenges the idea of a universal neurocognitive profile that could predict all types of addictive behaviors. The research found that while reward-related neurocognitive processes play a role in various non-substance-related addictive behaviors, these processes manifest differently depending on the specific behavior in question. The study suggests that instead of a single set of neurocognitive functions driving all addictions, there may be unique mechanisms at play depending on the behavior.
Addiction research has long focused on the neurocognitive mechanisms that underlie substance use disorders like alcohol or drug addiction. However, there has been far less investigation into the cognitive processes that might predict non-substance-related addictive behaviors, such as addictive eating, problematic pornography use, and excessive internet use.
Previous studies often suggested that impairments in cognitive control and reward processing might be common predictors across different types of addictive behaviors. However, these studies have often produced inconsistent results, particularly when looking at non-substance-related addictions.
In a previous cross-sectional study, published in Addictive Behaviors, the researchers found that different addictive behaviors were associated with unique cognitive and personality traits. By adopting a longitudinal approach in their new study, they aimed to clarify whether specific neurocognitive functions could predict changes in addictive behaviors over time.
“Neurocognition and addictive behaviors can influence each other in a bi-directional manner. For instance, impaired neurocognitive function can increase the likelihood of developing and maintaining addictive behaviours, while chronic engagement in such behaviours can also modulate or impair neurocognitive function,” explained study author Erynn Christensen, a postdoctoral researcher affiliated with the Feinstein Institutes for Medical Research and Zucker Hillside Hospital.
“Cross-sectional studies, which measure both variables at the same time, cannot determine the direction of their relationship. Longitudinal models, however, allow us to investigate how one variable can predict another variable, thus clarifying the direction of the relationship.”
The researchers recruited 400 young adults aged 18 to 35 years from the general population, with the aim of capturing a diverse sample that reflects the typical onset age for many addictive behaviors. The participants were primarily Australian residents and were free from any neurological or psychotic disorders, ensuring that the study focused on neurocognitive functions relevant to addiction without interference from other significant health conditions.
The study involved a series of neurocognitive tasks and self-report surveys administered at three points: baseline, three months, and six months. These tasks were delivered online and were designed to assess various cognitive functions that have been previously linked to addiction, such as response inhibition, reward learning, and decision-making under uncertainty.
For example, one task measured how well participants could inhibit their responses to a visual cue, reflecting their cognitive control, while another assessed their tendency to take risks when faced with uncertain outcomes, a measure related to reward processing. Additionally, participants completed surveys that measured their engagement in different addictive behaviors, allowing the researchers to track changes over time.
The researchers found that no single neurocognitive function consistently predicted all types of addictive behaviors, challenging the idea of a universal cognitive profile for addiction. However, specific neurocognitive functions were linked to certain behaviors.
“We showed that reward-related neurocognitive processes were implicated across each non-substance addictive behavior, but in different ways, underscoring their importance in driving addictive behaviors and making them an important focus of further research,” Christensen told PsyPost.
For example, participants who exhibited poorer performance monitoring (detecting and evaluating errors or discrepancies in one’s actions ) were more likely to develop addictive eating behaviors over the study period. This suggests that difficulties in monitoring one’s actions may play a role in the development of compulsive eating habits
In contrast, the study found that lower delay discounting, meaning a greater ability to delay gratification, was unexpectedly associated with higher levels of problematic pornography use over time. This counterintuitive finding suggests that the relationship between cognitive control and specific types of addiction may be more complex than previously thought.
“A finding that particularly surprised us was that a preference for delayed gratification predicted more problematic pornography use over time,” Christensen said. “We actually expected the opposite result because previous research has shown that individuals with problematic pornography use tend to have a higher preference for immediate gratification compared to those with regular, non-problematic use.”
“We suspect this discrepancy arises because our study examined problematic pornography use dimensionally, across a spectrum of low to high levels, while the other study used a categorical approach, comparing distinct groups based on severity, but it is difficult to say. We definitely need more research devoted to better understanding the neurocognitive mechanisms of problematic pornography use.”
Additionally, the study found that greater sensitivity to reward-related cues predicted more severe addictive eating behaviors at later stages, indicating that individuals who are more easily distracted by high-value rewards may be more vulnerable to developing compulsive eating patterns.
“To date, research on the neurocognitive predictors of non-substance addictive behaviors, such as problematic eating, problematic pornography use, and problematic internet use, is extremely limited,” Christensen told PsyPost. “Furthermore, there has been little to no comparison of the independent drivers behind these behaviors. Current neurocognitive models of these addictive behaviors often assume that common functions are impaired across these behaviors and that these impairments can predict poorer outcomes.”
“However, our findings suggest this may not be the case. We did not identify a core set of specific neurocognitive functions that reliably predicted addictive behaviors across multiple behavior types, suggesting that there may be partially distinct neurocognitive mechanisms contributing to addiction, depending on the specific addictive behavior under consideration. These findings are cohesive with our previous work showing similar relationships cross-sectionally, as well as findings from a systematic review of the current state of the literature. However, more work should be done to interrogate these differences.”
While the study provides new insights into the neurocognitive mechanisms underlying different types of addictive behaviors, it also has limitations. One of the main limitations is that the study sample was drawn from the general population and exhibited relatively low levels of engagement in addictive behaviors.
“Our failure to identify common neurocognitive predictors across all addictive behavior types (alcohol, eating, internet use and pornography) does not mean they do not exist, they just might be more important at later stages of addictive behavior (e.g. when the behaviour reaches clinical levels), and could play a role in treatment success,” Christensen noted.
“Future work is needed to establish whether our findings hold in samples with more severe problematic use. However, a systematic review published by myself and the team last year revealed, when looking at later stages of addiction and treatment outcomes, we see neurocognition differentially predicted different substance use types and gambling.”
“Another important consideration is how we measured neurocognition in this study,” Christensen added. “Neurocognitive functions, such as cognitive control and reward-related processes, are dynamic and context-dependent, influenced by factors like sleep, stress, craving, and exposure to addiction-related cues, like walking past a bar. As a result, single-time-point assessments used in this study may not effectively capture these fluctuating processes.”
“Ecological Momentary Assessment (EMA) paradigms offer a promising solution. EMA involves multiple short neurocognitive assessments over time (e.g., daily or several times a day), providing frequent snapshots of neurocognition. This approach can track within-person fluctuations across various contexts, psychological states, and environmental conditions. I believe there is a lot of value in this approach and is the way forward for the field of neurocognitive assessment.”
The study, “A comprehensive evaluation of the neurocognitive predictors of problematic alcohol use, eating, pornography, and internet use: A 6-month longitudinal study,” was authored by Erynn Christensen, Lucy Albertella, Samuel R. Chamberlain, Chao Suo, Maja Brydevall, Jon E. Grant, Murat Yücel, and Rico Sze Chun Lee.
