The temporal structure of goal‐directed and habitual operant behavior

Thrailkill et al. (2024) asked a simple question: do goal-directed and habitual behaviors look different under a microscope? They trained rats to press a lever for food, then tested whether the tiny bursts of pressing—called bouts—changed after the food was devalued.

The team used a two-phase design. First, rats earned pellets on a variable-time schedule mixed with a fixed-ratio five. Later, the pellets were pre-fed or paired with illness to kill their value. The researchers then compared bout length, pause length, and response rate between valued and devalued conditions.

Despite clear drops in overall pressing after devaluation, the fine-grain timing stayed the same. Bout lengths, pauses, and early response speeds did not signal whether the behavior was goal-directed or habitual.

The result means temporal micro-structure alone cannot tell you which mental process drives the lever press. You need other evidence—like a devaluation test—to make that call.

Gildea et al. (2025) extends the story. They also tracked bouts but added space: rats had to walk back to the lever after each press. Their bouts still lengthened when the fixed-ratio grew, showing reinforcement rules—not just proximity—shape the burst. Thrailkill’s null finding now fits: if bout rules are set by the schedule, they will not suddenly shift just because the goal lost value.

Lemons et al. (2015) and Smith et al. (2014) used similar bout math. Both found that schedule tweaks (more FR5 requirements or adding a change-over delay) reliably stretched or shortened bursts. Thrailkill’s data say those same metrics stay flat after devaluation, drawing a clean line: reinforcement parameters control bout size; motivational state controls whether you bother to show up.

McSweeney et al. (1993) warned that averaged time patterns can fool you. Thrailkill heeded the warning by checking individual intervals and still found no difference, strengthening the null. Together, the papers tell us to stop hunting for a hidden temporal fingerprint and instead use proven tests like devaluation or extinction.

When you watch a client repeatedly zip through five picture cards, do not assume the pace tells you if the chain is habitual. Thrailkill’s work says pace is schedule-bound; value is tested separately. Pair your timing data with a quick reinforcer devaluation or brief extinction probe before you claim “it’s automatic now.” Save yourself from false positives and keep treatment decisions anchored to real motivation shifts, not micro-timing hunches.