Conclusions: OFC→M2 is necessary to translate cued reward memories into goal-appropriate instrumental rules.

Thank you to everyone that contributed to this work! @ronkeiflin.bsky.social

At test, reward-predictive cues biased rule selection, promoting the rule associated with the predicted outcome. Silencing the orbitofrontal-to-secondary motor pathway disrupted this cue-guided *rule* selection. Critically the same manipulation spared cue-evoked *action* selection in PIT paradigm.

Rats learned two different rules for completing a 3-step action sequence (delayed nonmatch to sample or visual rule), with each rule leading to a distinct reward outcome. www.keiflinlab.com/instrumental...

To bring this higher-order level of behavioral organization into the lab, we adapted the iconic Pavlovian-to-Instrumental Transfer (PIT) paradigm and developed the Pavlovian-to-Rules-to-Instrumental Transfer (sPRInT).

But real-world reward pursuit is rarely a choice between isolated actions. Different goals often require different strategies -or higher-order rules- that organize multiple actions into a coordinated sequence.

Classic decision-making tasks (in rodents) often reduce choice to two competing actions: press the left lever or press the right lever.

Conclusions: OFC→M2 necessary to translate cued reward memories into goal-appropriate instrumental rules.

Thank you to everyone who contributed to this work! @ronkeiflin.bsky.social

At test, reward-predictive cues biased rule selection, promoting the rule associated with the predicted outcome. Silencing the orbitofrontal-to-secondary motor pathway disrupted this cue-guided *rule* selection. Critically the same manipulation spared cue-evoked *action* selection in PIT paradigm.

Rats learned two different rules for completing a 3-step action sequence (delayed nonmatch to sample or visual rule), with each rule leading to a distinct reward outcome. www.keiflinlab.com/instrumental...

To bring this higher-order level of behavioral organization into the lab, we adapted the iconic Pavlovian-to-Instrumental Transfer (PIT) paradigm and developed the Pavlovian-to-Rules-to-Instrumental Transfer (sPRInT).

But real-world reward pursuit is rarely a choice between isolated actions. Different goals often require different strategies -or higher-order rules- that organize multiple actions into a coordinated sequence.

Classic decision-making tasks (in rodents) often reduce choice to two competing actions: press the left lever or press the right lever.