- Question
- Should scientific research focus on facial coding systems to explain what people are thinking?
- Position1 of 2›
- Yes, we should focus scientific research on facial coding systems to explain what people are thinking
- Argument‹2 of 2
Faces are already being read, and research is what would discipline it
The argument
The second case concedes that the science is unsettled and argues that this is the reason to pursue it rather than to leave it alone. Inference from faces to inner states is not a prospect awaiting a research programme. It happens constantly. Interviewers judge candidates on whether they seem sincere, clinicians on whether a patient seems to be in pain or concealing distress, border officers on whether a traveller seems nervous, juries on whether a witness seems to be lying. These judgements are made by people with no training, no error rate and no accountability, and they carry real consequences. Automated systems making the same claim are also being sold already, and their accuracy is asserted by the organisations selling them. Proponents argue that the only thing capable of constraining either practice is a body of research rigorous enough to establish what facial movement can and cannot support. Where a claimed inference is unfounded, it is research that shows this, and a published negative finding is the strongest instrument available against a vendor claim or a confident interviewer. Where it holds under some conditions and not others, only systematic work can mark the boundary. On this view, declining to fund the field does not prevent faces from being read; it leaves the reading to intuition and to commerce, neither of which publishes a method or accepts refutation. Proponents accept that research attention partly legitimises a field, and argue the trade is still worth making, because an unexamined practice already in use is in a worse position than a studied one. The purpose of focusing here is as much to establish the limits of what faces reveal as to extend them.
Premises
Counter-arguments
Critics answer that research attention does not reliably function as a brake. Funding a field builds laboratories, careers and commercial partnerships whose participants have an interest in the inference being tractable, and the literature that results is as likely to supply citations for vendors as to refute them. They add that the disciplining function does not require making this a priority: a small programme designed to test the core inference would establish its limits, whereas focusing research here expands capability at the same time. The choice presented as between study and neglect is really between a narrow validation effort and a field.
Rejecting the premises
[Rejecting P3] Establishing limits does not require making the field a research priority, since a narrow programme designed to test the core inference would do that while focusing effort here expands capability at the same time. [Rejecting P4] Published findings function as citations as readily as refutations, and a literature produced by laboratories with commercial partnerships is not a reliable brake on the vendors it is shared with. [Rejecting C] The choice is not between study and neglect but between a contained validation effort and a field, so the conclusion rests on an alternative that was never the only one available.