A functional analysis of mentalistic terms in human observers.

The author shows how to study what people say about behavior without using mind words. He calls it the Reno approach. You watch a pigeon work for food. You record every peck and hop. Then you ask a person to describe what the bird did. You treat the person's words as more behavior, not as clues to hidden thoughts.

The paper is a recipe, not an experiment. It lists the steps: set up the operant box, run the pigeon, record the observer's speech, graph both streams side-by-side. No numbers, no stats—just a clear way to turn 'He looked confused' into 'When the red light came on, she said confused and the bird paused 3 s.'

The method lets you replace mental labels with plain environmental events. Instead of 'The bird expected food,' you write 'Red light → 12 pecks → food hop → bird paced.' The observer's sentence becomes a report of past stimuli and responses, not a guess about a mind.

By graphing bird behavior and observer talk together, you see which bird events trigger mental words. The words pop out right after long pauses or errors. The pattern shows that 'expect' is evoked by extinction, not by a hidden brain state.

Morris et al. (1982) warned us to drop cognitive words from behavior analysis. This paper shows how to do it. Where K et al. argued, Grant (1989) demonstrates. One gives the rule, the other hands you the tool.

Spanoudis et al. (2011) and Aragona et al. (1975) ran similar pigeon setups but counted pecks to test matching law. Grant (1989) uses the same boxes and birds, yet asks a new question: What will a human say while she watches? The trio forms a ladder—first measure bird behavior, then describe it, then interpret the description.

Furrebøe et al. (2017) say behavioral economics should ditch mentalistic talk and use operant tools. This 1989 paper already showed the way. Frølich extends the argument to economic choice; S supplies the method.

When you write session notes, swap mind words for observable events. 'The client was frustrated' becomes 'After the third error, he shoved the table.' Graph the shove and the error together. Your team can see the contingency and change it. The Reno frame keeps treatment plans anchored to what we can see and count, not to what we guess the client feels.