[HN Gopher] Efficiently irrational: deciphering the riddle of hu...
___________________________________________________________________
Efficiently irrational: deciphering the riddle of human choice
[pdf]
Author : telotortium
Score : 21 points
Date : 2022-06-07 18:31 UTC (4 hours ago)
(HTM) web link (www.gwern.net)
(TXT) w3m dump (www.gwern.net)
| ouid wrote:
| Nick didn't lose 827 million dollars. Nick lost his job. Nick
| behaved rationally. Every choice was a choice between certainly
| losing his job, and maybe losing his job.
| telotortium wrote:
| Abstract:
|
| > For the past half-century, cognitive and social scientists have
| struggled with the irrationalities of human choice behavior;
| people consistently make choices that are logically inconsistent.
| Is human choice behavior evolutionarily adaptive or is it an
| inefficient patchwork of competing mechanisms? In this review, I
| present an interdisciplinary synthesis arguing for a novel
| interpretation: choice is efficiently irrational. Connecting
| findings across disciplines suggests that observed choice
| behavior reflects a precise optimization of the trade-off between
| the costs of increasing the precision of the choice mechanism and
| the declining benefits that come as precision increases. Under
| these constraints, a rationally imprecise strategy emerges that
| works toward optimal efficiency rather than toward optimal
| rationality. This approach rationalizes many of the puzzling
| inconsistencies of human choice behavior, explaining why these
| inconsistencies arise as an optimizing solution in biological
| choosers.
| nopinsight wrote:
| A hypothesis I came up with a few years ago:
|
| Stereotyping is a result of the brain optimizing for efficient
| use of limited data and compute. Humans often over-generalize and
| come up with possible/plausible though unnecessarily true
| inferences. (And since the mechanism is likely evolved,
| countering stereotyping requires active training and conscious
| awareness of the tendency).
|
| Note: I only read the introductory paragraph but it looks like
| the above hypothesis is related to the paper.
| panarky wrote:
| Sounds like the "System 1" portion of Daniel Kahneman's book
| "Thinking, Fast and Slow" from 2011.
|
| "System 1" is fast, reflexive, automatic, emotional,
| unconscious and easy. It's based on shortcuts and stereotypes,
| often gets things wrong.
|
| "System 2" is slow, deliberate, difficult, calculating, based
| on focused reasoning and cognition.
|
| Since "System 2" requires so much effort, most people spend
| most of their waking lives in "System 1".
|
| https://en.wikipedia.org/wiki/Daniel_Kahneman
| telotortium wrote:
| Need to do a full read-through, but part of what this paper
| seems to say is that the System 1 vs. System 2 dichotomy is
| not the best description of how we make decisions. From Page
| 5:
|
| > How can one make sense of such inconsistent behavior? One
| answer [16] drawn from the multiple-selves model is that our
| brains have two independent modules: an 'impulsive' chooser,
| which only values immediately available rewards; and a
| 'rational' patient chooser, which values all rewards as a
| function of the delay-to-reward duration. In this conception,
| human behavior represents the combined action of these two
| modules, each internally consistent when acting alone, but
| yielding strongly intransitive behavior when combined.
|
| and page 7:
|
| > Could it be that there is a unitary representation of a
| single utility-like signal, rather than the many competing
| signals predicted by multiple-selves theory?
|
| > At the same time, a similar examination of another
| behavioral pattern sometimes seen as an inconsistency, loss
| aversion, yielded similar results [47]. Humans often treat
| losses relative to the reference point as more consequential
| than equivalent gains, presumably because losses are more
| anxiety-provoking. In this study, brain activity was examined
| as decision-makers evaluated financial lotteries that
| involved varying levels of gains and losses. As in
| intertemporal choice, these data revealed that human choosers
| were inconsistent, but that this inconsistency did not appear
| to arise from multiple modules, one only anxious and one only
| logical. Instead, the study revealed that several brain areas
| irrationally encoded losses and gains asymmetrically, but not
| by using multiple independent modules (but see [48] for one
| counter-example).
|
| > A series of three subsequent meta-studies strongly support
| this conclusion that a single decisionvariable is represented
| neurally and that emotional and rational evaluations are
| fully integrated [49-51]. All three of these metastudies,
| together aggregating data from hundreds of studies, tell the
| same story. Value representations in the brain do not reflect
| multiple competing agents. Instead, these data suggest that
| the irrationalities we observe in behavior reflect a
| fundamental irrationality in the neural representation of
| subjective value, in humans, nonhuman primates [52], rodents
| [53], and perhaps even insects [54]. The representation of
| subjective value appears fundamentally imperfect, but
| imperfect in similar ways, in animals separated by hundreds
| of millions of years of evolution.
| derbOac wrote:
| The two-system model has been criticized for awhile now
| from several perspectives. This paper is in line with that
| trend.
| cinntaile wrote:
| Ignore the system I parts about priming though, they have had
| trouble replicating those studies.
___________________________________________________________________
(page generated 2022-06-07 23:02 UTC)