Science1 publisher2 min readPublished
Activating the striatum's indirect pathway halts cocaine-driven licking in mice
A Hebrew University team traced cocaine's compulsive licking in mice to the orofacial action-selection circuit of the ventrolateral striatum, then turned the behavior off and on by moving the region's two competing pathways.
The Scientist · Science desk
What happened
- PhD researchers Ben Jerry Gonzales and Itay Shalom, working under Ami Citri at the Hebrew University of Jerusalem, used repeated cocaine dosing in mice as an experimental model of behavioral inflexibility.
- The team built STEREO, a deep-learning video platform that classifies and tracks an animal's entire behavioral repertoire from raw high-speed footage instead of hand-scored single actions.
- By the fifth day of cocaine exposure, licking the floor and enclosure walls filled more than 60% of the animals' active time, a behavior virtually absent in drug-naive mice.
- Optogenetically activating indirect-pathway neurons in the ventrolateral striatum interrupted the stereotyped licking immediately, and the animals returned to varied exploratory actions.
Compiled by The ScientistSomething wrong?How this is made
Why it matters
- capability Tracking the whole repertoire converts the question from whether an animal licks into what share of its behavior licking displaced, and a model of narrowing flexibility needs that share.
- constraint Because the licking came back as soon as the light went off, what the indirect pathway controls is whether the loop is expressed at that moment, and the result says nothing about the drug state behind it.
- precedent Loss of behavioral flexibility is a shared hallmark across substance use disorders and other neuropsychiatric conditions, so a named subregion and pathway gives other compulsion models a specific coordinate to test.
Each of the two pathways was moved in both directions, and each direction gave the opposite result [19]. Silencing the indirect pathway lengthened the bouts of repetitive licking and impaired the animals' ability to switch tasks [12]. On the direct side, inhibition attenuated the cocaine-induced stereotypy [13], and activation produced rigid, repetitive behavior in animals that had not been given the drug [14]. That last arm is what supports the hijacking language, because the behavior appeared with no cocaine on board.
The region is a specialized one. The ventrolateral striatum is the basal ganglia subregion that controls fine movements of the mouth, tongue and face [8], and grooming and licking are among the natural actions it already selects [2]. Inside it, the direct and indirect pathways compete for control of motor execution [9]. Cocaine, on the Hebrew University account, does not write a new motor program into that competition. It takes over one that was already there [2].
Hand scoring is part of why this was hard to see. Behavioral neuroscience has typically relied on human scorers or on one isolated action at a time, which makes transitions across long sessions difficult to analyze [5]. "We wanted to capture behavior as an observer actually sees it: grooming, licking, exploring, but this was impossible to score manually," said co-lead author Itay Shalom [15]. "STEREO allowed us to track the entire behavioral repertoire progressively narrowed until one type of action came to dominate," Shalom said [16].
Rearing at the walls, sniffing and walking were the actions that gave way [7]. Once licking passes 60 percent of active time, everything else the animal might have done shares less than 40 percent of it [18].
The announcement leaves out the number of animals tested, where the work was published, and whether the manipulation changes cocaine seeking [20]. Optogenetic stimulation in a mouse identifies a circuit, and it is not a treatment [10]. These experiments also do not test whether compulsions built out of actions other than mouth movements run through the same push-pull [9].
What to watch
- A peer-reviewed paper with animal numbers and effect sizes, which would let the 60% figure be checked against a control distribution.
- Whether the same ventrolateral striatum push-pull changes cocaine self-administration or relapse, not only orofacial stereotypy.
- Whether other labs reproduce STEREO's repertoire classification, and whether it detects narrowing in compulsion models that do not involve licking.