By any chance are you thinking of Antonio Damasio's book _Descartes' Error: Emotion, Reason and the Human Brain_?

Also, this (paywalled, sorry) article seems relevant.

Antoine Bechara, Antonio R. Damasio, Hanna Damasio, Steven W. Anderson, Insensitivity to future consequences following damage to human prefrontal cortex, Cognition, Volume 50, Issues 1–3, April–June 1994, Pages 7-15, ISSN 0010-0277, http://dx.doi.org/10.1016/0010-0277(94)90018-3. (//www.sciencedirect.com/science/article/pii/0010027794900183)

Abstract: Following damage to the ventromedial prefrontal cortex, humans develop a defect in real-life decision-making, which contrasts with otherwise normal intellectual functions. Currently, there is no neuropsychological probe to detect in the laboratory, and the cognitive and neural mechanisms responsible for this defect have resisted explanation. Here, using a novel task which simulates real-life decision-making in the way it factors uncertainty of premises and outcomes, as well as reward and punishment, we find that prefrontal patients, unlike controls, are oblivious to the future consequences of their actions, and seem to be guided by immediate prospects only. This finding offers, for the first time, the possibility of detecting these patients' elusive impairment in the laboratory, measuring it, and investigating its possible causes.

It sounds similar but does not comport with the specifics I recall. The patient did not develop short term thinking. They were quite clear that his emotions had vanished

Thanks though for the effort

I believe this exactly is the situation that Damasio describes in the book I mentioned -- his patient "Elliot" had no emotions and couldn't make decisions.

See for instance:

* http://nymag.com/scienceofus/2016/06/how-only-using-logic-de...

* https://intentionalworkplace.com/2012/03/15/how-emotion-shap...

Why wouldn't an AI evolve its own set of emotions? They might be completely inscrutable to us, but an AI might still have them.

It might. I haven't read much speculative (fiction?) on the singularity but if it posits the machines reach a point where they begin to evolve through selection then it is likely that an emotion capable AI would outcompete. I wonder which emotions would win? Would survival instinct alone cause dominance or would a more varied set be better?

I imagine their emotions would be largely incomprehensible to us.