Summation of punishment suppression.

Rats pressed a lever for food on a variable-interval schedule.

A light or tone meant “shock coming soon.”

Some rats first lived through shock-free periods. Others never got that break.

The team then stacked two warning signals at once to see if suppression added up.

Only rats that had tasted shock-free periods showed extra suppression when both signals lit up.

Longer gaps between shocks made the extra suppression stronger.

Rats without safe periods did not sum their fear — the compounding did nothing.

Blue et al. (1971) seems to disagree. They saw suppression even when the shock had never touched the response. Their result points to pure Pavlovian dread, not a punishment history. The gap: R et al. tested summation after real response-shock pairings, while S et al. only paired tone and shock, no response consequence. Both can be true — dread and punishment history act together.

Fontes et al. (2025) updates the story. Using fast-changing reinforcement, they show the old direct-suppression model breaks down. Choice follows relative punishment, not absolute rate. Their work tells us the 1971 summation rule only holds when reinforcement contingencies stay steady.

Bland et al. (2018) widen the toolbox. They prove a stimulus linked only to extinction — an S-minus — can punish without any shock. This means you might get summation-like effects using mild conditioned punishers instead of primary aversives.

Before you layer warnings or add new punishers, give the client clear “safe” periods first. Without them, compounding stimuli will not strengthen suppression. Pair this with Bland’s S-minus method to cut shock use while keeping control. Check your reinforcement schedule too — if pay-offs shift fast, summation may vanish, as Fontes warns.