Reggie the Raccoon: A User's Manual

 Or What Happens When a Brain Optimised for Threat Detection Has to Make Small Talk

The therapist's name was Dr. Pamela Osei, and she had, in fourteen years of clinical practice, treated executives with god complexes, founders who had mistaken their net worth for their self-worth, and one memorable hedge fund manager who had become convinced that his succulent collection was trying to communicate with him. She had not, until now, treated a raccoon.

"I don't need to be here," Reggie said, from the corner of the couch furthest from the door, which he had assessed upon entry as the optimal defensive position.

"Most people say that," said Dr. Osei.

"Most people are not here because their manager wrote 'struggles with interpersonal calibration' on a performance review and HR interpreted that as a therapy referral."

"And what do you think it means?"

Reggie considered this with the same methodical patience he applied to anomaly detection pipelines. "It means," he said, "that I told Graham from Internal Audit that his sampling methodology was, and I want to be precise here, epistemically indefensible. And Graham cried. Which I did not anticipate. I had modelled several responses. Crying was not in the top five."

Dr. Osei wrote something in her notebook. Reggie noted she used a pen, which he found both quaint and statistically inefficient.

Here is what the notebook would not capture, because it cannot be observed from a leather chair in a consulting room in Farringdon: Reggie's brain is not broken. It is differently optimised. This distinction matters enormously, and the fact that it took the psychiatric community the better part of a century to arrive at it is, depending on your disposition, either a triumph of incremental science or a staggering indictment of how long it takes institutions to update their priors when the data has been sitting right there.

What Reggie has, though he would object to the framing of "has," preferring "operates via", sits somewhere on the landscape that researchers now describe with considerably more nuance than they did when the DSM still classified left-handedness as a developmental concern. The current understanding of autism spectrum conditions is that they represent not a deficiency but a different cognitive architecture: one characterised, in many cases, by heightened pattern recognition, strong systemising tendencies, reduced interest in social hierarchies, and a nervous system that processes sensory information at a volume most people cannot imagine and nobody warned you about.

The social bit is not, as it is sometimes popularly described, an absence of empathy. It is, more accurately, a different channel. Reggie does not read faces well. He reads data extraordinarily well. These are not opposites. They are, in the current research literature, increasingly understood as the same underlying trait expressing itself differently depending on what kind of signal is in front of you. Simon Baron-Cohen at Cambridge has spent decades on this. The raccoon community at Morridian Fins has spent approximately fourteen months noticing that Reggie is better at spotting a fraudulent invoice than anyone in the building and considerably worse at noticing when someone is being passive-aggressive in a meeting, which, for the record, is most of the time.

"Tell me about your work," said Dr. Osei, because therapists, like good anomaly detection systems, build a baseline before they flag anything.

Reggie told her about his work. He told her for twenty-three minutes. He covered Benford's Law, the Henderson Situation, the inadequacy of rule-based monitoring, the specific genius and frequent misapplication of SMOTE, and the deeply personal grievance he carried regarding ROC-AUC being used as a primary evaluation metric on imbalanced datasets. Dr. Osei did not understand approximately sixty percent of this. She noted, however, that he was not performing. He was not managing an impression. He was simply telling her the truth about what he found interesting, with the complete absence of social filtering that most adults spend enormous energy maintaining.

This is, the literature would tell you, both a feature and the reason for the performance review.

The ADHD component, and it is, in Reggie's case, present, because these things tend to travel together in what clinicians call comorbidity and what Reggie calls "the full stack", adds a particular texture. His attention is not deficient. It is selective in ways that do not align with what open-plan offices were designed to accommodate. Give him a spreadsheet with anomalous digit distributions and he will not surface for four hours. Ask him to sit in a project update meeting where someone is reading slides aloud that everyone in the room has already read, and his nervous system will begin generating a low-level distress signal that he has learned, over time, to manage by writing chi-squared test variations in the margins of his notepad. The dopamine system in ADHD brains is not absent. It is calibrated differently: spiking hard on novelty, pattern resolution, and the specific satisfaction of finding the thing that was hidden. Flat-lining on administrative process and small talk about the weekend.

This is not laziness. This is neuroscience. Russell Barkley has been saying this since the nineties. It does not stop people from writing "could engage more proactively in team meetings" on performance reviews.

"Do you find it difficult," Dr. Osei asked, toward the end of the session, "when people don't respond the way you expect?"

Reggie was quiet for a moment. Outside, London was doing what London does, which is to produce an unrelenting stream of noise and signal that most people filter unconsciously and that Reggie filters manually, at cost.

"I don't always know what to expect," he said, finally. "With people. The variables are not documented. There's no schema. You can't run a query." He paused. "Data doesn't lie to you. Data doesn't get offended. Data doesn't say it's fine when it isn't fine and then bring it up six weeks later in a completely unrelated context."

Dr. Osei found, to her mild surprise, that she did not have an immediate response to this.

"That sounds," she said, carefully, "quite lonely."

Reggie looked at the window. Somewhere beneath them, the city was generating approximately eighty million network flow records per day, each one a small truth, each one waiting for someone patient enough to look.

"It's fine," he said.

His pager went off.

He was already standing.

Dr. Osei has noted that Reggie attended the session, engaged substantively, and demonstrated considerable self-awareness. She has recommended a follow-up. Reggie has it in his calendar. Whether he will remember that calendars require attendance as well as entries is, at this stage, a known unknown.

The shortbread in the waiting room was from Marks & Spencer. Reggie rated it a 6.4 out of 10. He has a spreadsheet.

References

Baron-Cohen, S. (2020) The Pattern Seekers: How Autism Drives Human Invention. New York: Basic Books.

Barkley, R.A. (1997a) ADHD and the Nature of Self-Control. New York: Guilford Press.