Lie to Me

This post was inspired by Amy Blaschka and her post "Why People Tell Me Things" and the show "Lie to Me", which Amy is a fan of.  In spite of the brilliance of Paul Ekman, on whom the show is based, his works do not address the question as to why we fall for lies, so I thought I'd give a little more background in how lying works besides saying we are not good at recognizing the signs - which is Paul Edman's explanation.

Researchers Sophie Snowden and her collaborators at King's College London have done a very fine experiment to get at the causes of lie in-detection.  First, they were able to negatively correlate world view, opinion, biases with lie detection, and then to suppress functioning in the parts of the brain where we make judgments about the world with an attendant improvement in lie detection. Before we explain the experiment let's set the stage by explaining how the brain makes coherent world views, then how to interfere with the brain's functioning using electromagnetic waves so that we can properly think about this beautiful experimental result.

The experiment consisted of two parts 1) story alignment between the liar and the lie (in-)detector, and 2) disabling part of the brain thought to be involved with how we postulate the reasoning of others through trans-cranial stimulation. In the first part, there was a mild effect for dis-aligning a story's context between people but a large effect when the “social judgement” part of the brain was disabled. The social judgement disabling is labeled “TPJ” below.

As you can see the big improvement in lie detection happened when the Temporal and Parietal Junction (TPJ) was disabled through trans-cranial stimulation.

 The ability to detect a lie improved from barely above chance to 60% by disabling part of the brain! Which must mean - we are hardwired, or at least programmed in part of the brain, to accept lies? Or does it? The difference between an elegant experiment and a detailed experiment will become apparent as we investigate this phenomenon. To set the stage we have to talk for a moment about what an experiment is, then we have to explain a bit about brain science, before we come back to this fascinating result.

We learn in school that an experiment is designed to test an hypothesis. But, there are several factors at play that are not drawn out in most researchers education. For instance, the measurement tools may not be testing the hypothesis but measuring something else instead. Let's call this something else the unexplored hypothesis. And each hypothesis sits in a larger context called a model. If you accept the model, then the experiment, with proper tools, can test the hypothesis. Conversely, if you reject the model, then the experiment is a test of another unstated hypothesis. Here, we have a hundred or so year old model of the brain that says we are wired together like a telephone switch network.

Signals are sent from one area of the brain to another and processed a little bit at each step and then sent, like a telephone signal, to another area of the brain for further processing. In this model disabling the junction between the Temporal and Parietal regions of the brain, if they are critical to estimating the actions of others, is a nice binary test of lie detection. All you have to show is that the TPJ is a social organ, that lying is social, and that when the TPJ is switched off it affects lying (or lie detection). So if you think we are a switch network, then breaking a link in the chain of reasoning means a broken brain function breaks the signal. That is, if we are unable to listen to our internal dialog and speculate about the actions of others then we are free to pay attention to the intentions of others, or so Dr Snowden's reasoning goes.

When the brain is stimulated from outside the cranium, using electricity, or magnetism, it forces a general release of neurotransmitters over an entire region. The smallest region of the brain we can affect from outside the cranium is about the size of a dime. The nerves in that region are then unable to further process information because all nerves are transmitting at once – the switchboard has had a denial of service attack and is overwhelmed. The results of the experiment are incontrovertible, whatever the TPJ does, disabling it improves lie detection. The only question is what does the TPJ do?

 For the Sophie Snowden, and her collaborators, the TPJ is involved in how we view the intentions of others and our sense of self. They site several papers that identify psychological characteristics associated with damage to the TPJ and these show that people with damage to the TPJ are prone to out-of-body experiences, social miscuing and so on. This fits rather nicely with the earlier alignment experiment showing that when intentions are misaligned it's harder for us to detect a lie, because our brain tries to make sense of the lie using the TPJ. When it is disabled we are less concerned withour own internal intentions and can focus on the intentions of others thinks Dr. Snowden. But we might just fall back on other better cues like a lag in response time, cues that are suppressed when we listen to our own internal tasks.

So far so good. But there are two untested hypothesis. 1) is that really what the TPJ does and 2) is the switchboard model any good at all. Let's look at 1) first. What does the TPJ really do?

Is it a social organ?

Or is it a a more general contextual organ?

Researcher R McKnell Carter proposes that it is a general contextual organ and, as such, is a necessary element of social judgement.  There is good evidence that the TPJ is subdivided into smaller regions with more specific, but similar functions, as reported by Hiroshi Horiguchi, and fellow researchers, at Stanford.

 They found a particular subregion of the TPJ that has a strong visual component and is activated by strong and intermittent light in a fashion similar to the Optic Tectum of the zebrafish, and the superior colliculus of humans. I know, why do we call it the Tectum in fish and the Superior Colliculus in us? Just to make thinking about the brain harder. Ugh... Anyway this result shows that, at least a part of, the TPJ is involved in visual contextualization, not just social contextualization and that it works in conjunction with the visual system.  In zebrafish the Optic Tectum also responds to flashes and helps the fish figure out it the flashes from a predator fish is about to make a meal of it.  Our TPJ appears to have a similar function.

Fish have an Optic Tectum but not a TPJ, which is specific to mammals. And this makes sense, since we can view the brain as having redundant layers that can reinterpret data in larger contexts. Mammals in general, and humans in particular, can add lots of information to a signal by using many parts of the brain in conjunction. Remember to hold onto that thought about “in conjunction,” because we need to understand how information is co-ordinated across distant regions of the brain. But back to the problem at hand. All the studies sited by Dr Snowden, and the experiment itself, affect large regions of the TPJ and so it might be true that one area of the TPJ is involved in social context processing specifically, and another in visual context processing, and another in auditory context processing, and so on, or they might all be mangled together. But most research points to specificity, for when the TPJ is damaged or inhibited, the effects are often cross channel problems - like hallucinations, synesthesia and so on. It also turns out that people with mirror-touch synesthesia have less brain volume in the TPJ. And we will have to remember to come back to what “brain-volume” means in a moment. In sort the TPJ helps us sort out what goes where and helps us to provide context. When it is damaged we fail to keep things straight. If it co-mingled signals then it we would not cross wires when it fails to function.

Dr. Snowden's hypothesis was that stimulation of the TPJ would suppress our own context and allow us to focus on the context of the other person. And that may indeed be true, but it could equally be true that without social context, we relied on other areas of the brain to form an opinion. Or phrased another way, when the context area of the brain is active it overrides other, more relevant, cues. In short, we suppress information that does not conform to our own mental model at a neurological level.

Lying to myself

Thus it turns out - I am always “lying” to me! Or at least my biases are reinterpreting real information and discounting whatever doesn't support my own hypothesis.

Now we arrive at our second unstated hypothesis – maybe the brain is not wired together like a switchboard but operates like group of radio transmitters sending out radio waves to distant parts of the brain. Different activities are co-ordinated at different frequencies and will “sync-up” when activated into a cohesive, reprogrammable, system. Sure, we are wired like a switchboard at one level, my hands are connected to my brain, but this design is not flexible enough to work on novel problems or even quick pattern recognition problems. To do that we need to process in parallel and that means we use waves to communicate across the brain, even when those parts of the brain are not directly connected. Researchers who have explored this possibility have been able to show that, because the waves add and subtract, like ripples in a pond, they are able to control neuron excitation.

Where the waves cross they add, or subtract, the chemical and electrical signals, so that when waves are in phase they have enough energy to let the neuron fire. When they are out of phase there is not enough energy to let the neurons fire. You can see from the above photo that the wavelengths and location of the source are powerful selectors of different areas of brain activation.

 Interestingly the high frequency gamma EEG waves, originate in the Optic Tectum and are taken up by the TPJ! And we remember from physics that the he higher the frequency the smaller the spacial resolution of incident waves, and the greater activation of brain specificity. There are much higher frequencies in the brain than we can measure trans-cranially using an EEG, so we have no real idea of how specific we can be in activating a small region of the brain. And what supports the brain wave function?  It's the white matter glial cells of the brain.  If the white matter mass is too small (in the TPJ) then it will not properly participate in the radio like system and social judgements and planning will be impaired.  This nicely ties in with our earlier observation about TPJ reduced brain volume.

The TPJ is activated 100 milliseconds after the Optic Tectum. In this paper Joy Geng points our the role of the TPJ in spatial neglect. That is, signals come to the TPJ and the TPJ is responsible for suppressing unimportant signals.

If TPJ is to send a fast-latency “interrupt” signal to the dorsal network that reorients attention, it logically follows that TPJ must generate an earlier output signal in response to sensory inputs than the dorsal system regions. However, the evidence from electrophysiology and TMS suggests that Fast Eye Field (FEF in the above diagram) actually responds earlier, not later, to sensory information than [the] TPJ

Not stated in the above quote is that the Optic Tectum activates even before the FEF. In any case it is clear that TPJ gamma activity begins 100 ms after direct electrical stimulation of occipital cortex. But if this is so, how can it suppress signals that have already been activated? This is because the TPJ must coordinate signals relevant to the current task. And this is done by having all the task related parts of the brain communicating on the same frequency. Any nerves outside of the active group do not even fire. They are not considered in the evaluation process. This is a good thing because you wouldn't want pay attention to the motions of the distant clouds while the lion is attacking. And in like manner you can not pay attention to the social cues that do not fit into your intentions. There is no way you can hear the miscue associated with lying.

So you leave the door wide open for other's to lie to you, and you to yourself.

This post is dedicated to Paul Drury who went to bat for the Blog Poets and is one of the reasons I am still writing for LinkedIn.