The science lecture. What research actually says about rumination, why worry feels productive when it isn't, and the one place the popular story runs ahead of the evidence
1 March 2027
Where. Outdoors, venue posted by the Friday before. Opens with the ten-minute detox.
This is the closest thing to a theory lecture in the course, so it is doing two jobs. Telling you what is known, and being honest about where the popular version of this material outruns the research. The second job matters more.
Reflection and rumination are not the same thing#
Susan Nolen-Hoeksema’s work from the 1990s onward established rumination as a distinct thing worth measuring. Repetitive, passive, self-focused thought about your distress and its causes, which predicts worse depressive outcomes rather than better ones. The key finding is not “thinking about your feelings is bad.” It is that a particular mode of thinking about them, circling the problem and rehearsing the same framing, behaves very differently from thinking that moves toward something.
Which is why the useful question in this course is never “am I thinking too much.” It is “is this loop adding anything, and would I notice if it stopped.”
The thing that makes it hard to quit#
Here is the counterintuitive part, and it is a theory rather than a settled fact, so treat it accordingly.
Thomas Borkovec’s cognitive avoidance model of worry proposes that worry is mostly verbal, words and sentences in your head, and that this is the point. Verbal worry is comparatively bloodless. It sidesteps the vivid mental imagery and the physical response that a feeling actually needs in order to be processed. On this account worry is not a failure to cope. It is a very effective short-term way of not having the feeling, which is exactly why it is self-sustaining. It feels like work because it is work, just not the work that would resolve anything.
Two honest caveats.
| Claim | Status |
|---|---|
| Worry is largely verbal and dampens emotional arousal | Well supported |
| Worry functions to avoid emotional processing | One leading theory (Borkovec), with rivals. Newman and Llera’s contrast-avoidance model reads the same data differently |
| Therefore unprocessed emotion is the cause of overthinking | Overreach. This model was built to explain clinical worry and generalised anxiety, not overthinking in general |
That last row is worth dwelling on, because “overthinking is just unprocessed emotion” is a genuinely appealing sentence and it claims more than the research supports. It is a good hypothesis to hold loosely and test on yourself. It is not a finding.
Venting is not the mechanism#
If unprocessed feeling were the whole story, the fix would be to let it out, and the evidence says that is not quite it either.
James Pennebaker’s expressive-writing paradigm (from 1986 onward) is one of the most replicated findings in the area. Writing about a difficult experience for fifteen or twenty minutes across a few days produces measurable improvements. But when researchers looked at which writing helped, the predictor was not emotional intensity. It was change in the language over the sessions. More causal and insight words, because, realise, understand. People building a coherent account rather than re-feeling the same thing.
So processing is a construction job, not a release valve. Which, unhelpfully for anyone hoping otherwise, means it takes about as long as it takes.
What to argue with#
Bring one of these to the discussion.
- If rumination is defined partly by being unproductive, is “unproductive rumination predicts bad outcomes” a finding or a definition?
- Every study here measures self-reported thinking. What would you expect that to systematically get wrong?
- The course’s whole premise is that you can catch yourself mid-loop. Is there any reason to think noticing is easier than stopping?
Teaching team
- Marisol Quaye, convenor