Object category learning: How long-term practice rewires the brain
Source PublicationJournal of Cognitive Neuroscience
Primary AuthorsCox, Scholl, Laws et al.
"Learning to categorise objects is like learning to drive a manual car. At first, you must consciously think about every gear change, using all your focus. After months of practice, your hands and feet just know what to do automatically, freeing your mind to listen to the radio."

Mapping the cognitive genome
The study claims that prolonged practice fundamentally shifts object category learning from a conscious effort to an automatic reflex. Yet, mapping this cognitive genome has historically been difficult. For years, researchers only observed the short-term effects of learning. They mapped brief reactions rather than long-term biological changes. This approach left massive gaps in our understanding of how human behaviour adapts over months or years.
The limits of biological mapping
To understand the difficulty of mapping complex biological networks, consider how scientists sequence DNA. Historically, mapping a genome relied on gene markers. These markers act as specific signposts, pointing to known traits. However, this method has blind spots. It only finds what it is looking for, missing the wider context. An alternative approach analyses GC content—the ratio of guanine and cytosine building blocks across a DNA sequence. Scanning for GC content reveals broad, structural patterns rather than specific, known genes. It is highly efficient for spotting overall regulatory regions but lacks the precise detail of individual gene markers. Evaluating the brain faces a similar divide. Short-term scans act like narrow markers, while long-term studies reveal broader structural shifts.
Object category learning over time
Most previous research on object category learning relied on brief tests lasting only a few hours. These short experiments consistently showed that the prefrontal cortex actively managed the new information. This created a frontal bottleneck. Because the prefrontal cortex was busy, it limited the ability of participants to multitask. Critics often argue that these short studies fail to reflect real-world learning. The new study addresses this flaw by using a long-term method. Researchers trained participants to categorise novel car images over five to ten weeks. Participants completed more than 30,000 trials on a mobile phone programme. The team then measured brain activity using fMRI and EEG rapid adaptation techniques. They compared the initial learning phase against the final results.
Bypassing the bottleneck
The converging fMRI and EEG results highlight a clear physical change in the brain. After extensive training, the visual areas of the brain became highly selective for specific categories, rather than just shapes. At the same time, connectivity between these visual areas and the prefrontal cortex decreased. Instead, the visual areas built stronger links directly to motor output centres. The study measured improved performance during dual-tasking exercises. This suggests that extensive experience allows the brain to bypass the frontal bottleneck entirely. The data could mean that when a task becomes truly automatic, the brain no longer requires flexible, conscious control. It simply reacts.