Being so important to identify deception, we are really bad at it, so we developed the equivalent of an intelligence network that would pass along information and evidence, thus rendering the need for an individual lie detector moot

Nonverbal Communication: Evolution and Today. Mark G. Frank, Anne Solbu. In: Social Intelligence and Nonverbal Communication pp 119-162, January 26 2020. https://link.springer.com/chapter/10.1007/978-3-030-34964-6_5

Abstract: One aspect of social intelligence is the ability to identify when others are being deceptive. It would seem that individuals who were bestowed with such an ability to recognize honest signals of emotion, particularly when attempts to suppress them are made, would have a reproductive advantage over others without it. Yet the research literature suggests that on average people are good at detecting only overt manifestations of these signals. We argue instead that our evolution as a social species living in groups permitted discovery of deceptive incidents due to the factual evidence of the deception transmitted verbally through social connections. Thus the same principles that pressed for our evolution as a cooperative social species enabled us to develop the equivalent of an intelligence network that would pass along information and evidence, thus rendering a press for an individual lie detector moot.

Keywords: Deception detection Evolution Emotions Behavioral signals Social life

Conclusion
Taken together, it is clear that there are strong signals for various emotions and intentions and a strong rationale for why these signals would be
‘engineered’ to solve a recurrent problem. And despite being wired to
detect these signals, humans are poor detectors of these signals once they
become subtle through efforts to conceal them. Yet this ability to spot
these dishonest and/or subtle versions of the signals would seem to be of
great benefit to any given individual in his or her quest to survive and
pass on his or her genes to the next generation. This sense that evolution
did not bestow our species with these internal event detectors seems puzzling, until we unpack some of the social structures of the ancient world.
It seems the cooperative structures, and little (at least initially) opportunities to ‘cheat’, often may have allowed, in essence, an intelligence network to be developed where pejorative information could be passed along
easily and cheaply to identify any particular cheater. Thus, the evolution
of cooperative behavior was the key to lie-catching. It seems logical that
there would be no strong independent press to develop internal cheater
detectors, when a strong social network would do the job for at a greatly
reduced cost (Smith, 2010).
Importantly, lie detection in the laboratory or in single case studies
does not fully translate to the real world, where gossip and relationships
with others matter (Haidt, 2001). People rely on gossip, even when accuracy may be limited (Sommerfeld, Krambeck, & Milinski, 2008); it may
nevertheless actually improve lie detection (Klein & Epley, 2015).
Moreover, it is through the influence from others that we may decide to
override our tendency to cooperate (Bear & Rand, 2016) and employ
conscious deliberation to make our decisions (Haidt, 2001). The alignment of emotions through empathy, and increased goal sharing (Tomasello,
Carpenter, Call, Behne, & Moll, 2005), as evidenced by the #MeToo
movement (Rodino-Colocino, 2018), gave rise to the same powerful
group thinking and sociality as seen in the emergence of human morality
(Jensen, Vaish, & Schmidt, 2014). Haidt (2001) states “A group of judges
independently seeking truth is unlikely to reach an effective consensus,
but a group of people linked together in a large web of mutual influence
may eventually settle into a stable configuration” (p. 826). This becomes,
functionally, a long-range radar type system that has agents reporting
back actions, behaviors, and relationships to each other, which in turn sets
the groundwork for recognizing inconsistencies regarding people not
being where they say they are, people being with people they deny knowing, and so forth. The presence of this communication network would
reduce the need to make individuals hyper-vigilant in every interaction,
or to individually develop super-acute deception detection skills. Likewise,
unusual interpersonal behaviors can trigger individuals to search for evidence to verify their hypotheses about someone’s veracity, and they can
then activate their social networks to verify the information provided by
the unusually behaving person (Novotny et al., 2018). These networks are
not just passive providers of information. Thus, the socially intelligent
person is the one who has the best access to the collective intelligence—
and likely the most friends, as believed by the Ugandans (Wober, 1974).
We believe the research literature has neglected this larger system in which
our social structures exist, which often detect the deception for us. Even
as our society expands, social media and movements like #MeToo have
become like the global village, where previously unacquainted individuals
can now verify the truth or falsity of each other, thus (hopefully) betraying
the attempted liar.