Caffeine Fails to Sharpen Visual Search, Distraction Resistance, or Attentional Learning in New Study

A single 200-milligram dose of caffeine did almost nothing to improve the way people search visual scenes, ignore distracting stimuli, or learn the statistical regularities of a task, according to a new study published in Cognitive Research: Principles and Implications. The only measurable benefit emerged in one narrow measure: the speed with which participants could recognize and report a target once they had found it in a demanding visual search task. Everything else, from filtering out non-targets to resisting a salient colored distractor to trading off mental effort against physical effort, was untouched by the drug.

The research, conducted by Sojung Youn, Jenna Glotfelty, Ming-Ray Liao and Brian A. Anderson at Texas A&M University, set out to probe corners of attention that earlier caffeine studies had largely skipped. Prior work had shown that caffeine can speed responses on tasks such as the Attention Networks Test and reduce Stroop interference, but the efficiency of attentional selection during visual search, and whether that efficiency can improve through learning, had not been examined under acute caffeine administration. The team designed three experiments to fill that gap, each comparing a caffeine group against a placebo group in a single-blind between-subjects design.

In the first experiment, 91 participants aged 18 to 35, recruited from the Texas A&M community, searched for a sideways letter T among L-shaped distractors, with the number of items on screen varying from 5 to 50. Half received 200 milligrams of caffeine via two dissolvable sublingual strips, which the researchers chose because sublingual absorption is substantially faster than ingested caffeine; the other half received caffeine-free breath strips. Crucially, the target was rendered in one particular color on half of the trials, so participants who were sensitive to those color statistics should find the target faster when it appeared in the high-probability color. This design allowed the researchers to separate two components of performance: the search slope, reflecting how efficiently non-targets are rejected as display size grows, and the search intercept, reflecting how quickly a found target is identified and reported.

The results were striking in what they did and did not show. Color-based statistical learning exerted a powerful influence: participants were markedly faster to find the target in the frequent color, producing both a reduced search slope and a reduced intercept. Caffeine, by contrast, modulated only the intercept, with a main effect of participant group on intercept but no effect on slope and no interaction with the learning-dependent effects. In other words, caffeine appeared to speed the recognition and reporting of a target once selected, but it did not help observers reject non-targets more efficiently, nor did it amplify the benefits gained from learning the color regularities. Bayesian analyses supported a model containing only the target-color effect for slope, and a model with both target color and group effects for intercept, with moderate evidence against any interaction between the two factors.

The second experiment asked a different question: can caffeine help people resist distraction? A new sample of 86 participants completed an additional singleton task, searching for a uniquely shaped target while trying to ignore a color singleton distractor that appeared on a subset of trials. One distractor color was presented on 50 percent of distractor trials, making it a high-probability color that participants could learn to suppress more efficiently. The distractors robustly captured attention, with a large response-time cost when a distractor was present, and that cost was significantly attenuated for the high-probability color, replicating the well-documented role of selection history in distractor rejection. But caffeine played no role whatsoever. The main effect of participant group was not significant, and a Bayesian analysis allowed the researchers to confidently reject a model in which caffeine modulated the magnitude of attentional capture. Accuracy showed the same pattern: a cost from the distractor, no difference between high- and low-probability distractors in accuracy, and no influence of caffeine.

The third experiment shifted from performance to motivation, examining whether caffeine changes how much mental effort people are willing to expend. Eighty-eight new participants completed a task adapted from the Adaptive Choice Visual Search paradigm, in which two targets, one red and one blue, were hidden among dozens of colored squares. Because one color was always more numerous, searching through the less-prevalent color was easier, but reporting a target in one color triggered a much harder physical effort requirement: participants had to squeeze a hand dynamometer at 40 percent of their calibrated maximum force for three seconds, versus 10 percent for 200 milliseconds for the easier option. When attentional and physical demands conflicted, participants weighed the trade-off carefully, increasingly choosing the harder-to-find target as it became easier to locate, producing a significant interaction between color distribution and physical demand that directly replicated earlier work. Caffeine shifted none of these preferences. There was no main effect of group and no interaction between group and either attentional or physical demand.

The authors interpret the pattern as pointing to a limited role for caffeine in the attention system. The single significant effect, on search intercept in the demanding serial search task, is itself difficult to pin down: it could reflect faster matching of a selected stimulus against a target template held in memory, faster build-up of attentional priority before selection, quicker initiation of attentional processing at display onset, or simply faster manual responding. The researchers note that if the effect were purely motoric, a similar speeding would have been expected in the second experiment, where the evidence against a general speeding was only equivocal. They suggest that previously observed caffeine benefits on attention may be more strongly mediated by how efficiently task-relevant information is processed once selected, as in executive attention tasks, and less about the guidance of spatial attentional orienting, which was consistently unsupported in the present study.

The study also carries a set of limitations that the authors lay out explicitly. The 200-milligram dose matches amounts used in prior studies that produced significant effects, but more robust effects might have emerged at a higher dose. The placebo strips, while caffeine-free, contained active ingredients such as menthol that could have transient alerting properties, so they do not represent an entirely neutral baseline. Participants self-reported abstaining from caffeine for at least four hours before testing, which was not objectively verified, and typical daily intake was not measured beyond screening criteria. Effects might have been stronger in habitual users who were caffeine-deprived, in infrequent users, or in sleep-deprived individuals, though the study was designed to test enhancement in a rested, moderate-dose context rather than the masking of withdrawal or sleep loss. The single-blind design is another complication: the experimenter who unpackaged the strips knew their contents, and participants might have guessed their condition from the flavor of the strips, although the authors argue that such demand characteristics would have been expected to produce larger effects than the very limited ones observed.

For everyday decisions about caffeine as a cognitive enhancer, the findings are sobering. Participants who received caffeine were no more efficient at filtering non-targets, no better at resisting distraction by salient stimuli, no better at leveraging statistical regularities in either target guidance or distractor rejection, and no more willing to accept difficult attentional demands in exchange for reduced physical workload. The authors suggest that the learning-dependent benefits they observed may have been so robust that they created a ceiling effect, leaving no room for caffeine to improve upon them, and that subtler statistical regularities might be more susceptible to modulation. But at least under conditions with predictable, high-probability regularities, moderate doses of caffeine do not appear to enhance the ability to exploit them. For professions that rely on high-demand visual search, such as radiology, air traffic control, and security surveillance, the study argues against relying on caffeine for a temporary boost to selective attention, at least when the goal is genuine cognitive enhancement rather than counteracting the effects of sleep deprivation, which was not examined here.

Subject of Research: The effects of acute caffeine administration on visual search, resistance to distraction, and attentional learning in healthy adults.

Article Title: Investigating the effects of acute caffeine administration on visual search, resistance to distraction, and attentional learning

Article References: Youn, S., Glotfelty, J., Liao, M.-R., & Anderson, B. A. (2026). Investigating the effects of acute caffeine administration on visual search, resistance to distraction, and attentional learning. Cognitive Research: Principles and Implications, 11(1), Article 54. https://doi.org/10.1186/s41235-026-00760-8

Image Credits: AI Generated

DOI: 10.1186/s41235-026-00760-8

Keywords: caffeine, visual search, attention, attentional capture, statistical learning, distractor suppression, cognitive enhancement, selective attention, psychopharmacology, response time, mental effort, placebo-controlled study