JC#4 Neural Activity in Speech-Sensitive Auditory Cortex

[detta]

For next week's journal club I will be presenting
"Neural Activity in Speech-Sensitive Auditory Cortex During Silence", Hunter et al. PNAS Jan 3, 2006 http://www.pubmedcentral.gov/articlerender.fcgi?tool=pubmed&pubmedid=16371474

Along with this paper, I would also like to present
"Where the Imaginal Appears Real: A positron Emission Tomography Study of Auditory Hallucinations" Szechtman et al PNAS Feb 1998
http://www.pubmedcentral.gov/articlerender.fcgi?artid=19222

The papers are both pretty short and the more recent paper makes more sense in context of the older paper.

Auditory hallucinations are a major feature of schizophrenia and previous studies have shown that "imagined" hearing and hallucinations have similar activational patterns in the brain. The older paper tries to discern between the "imagined" hearing, hallucinations, and external hearing using PET scans. The more recent paper argues that there is a baseline activation of the anterior cingulate cortex (the area activated during hallucinations) that can be activated even when there is silence. They hypothesize that it is the changes from normal baseline activity that cause the auditory hallucinations.

 

[neurocomp2003]

nice find. Hopefully I'll have tim eto read this one.

 

[CosminaPrisma]

Haven't read that yet, but it looks very interesting. :)

 

[detta]

“Neural Activity in Speech-Sensitive Auditory Cortex During Silence”
Hunter, Elckhoff, Farrow, Wilkinson, & Woodruff PNAS; Jan 3 2006

The investigators of this paper are trying to discern variation in baseline activity of the auditory cortex of the brain that may help in understanding the spontaneous activation during hallucinations, hypnosis, drug intoxication, and types of psychosis. They hypothesize that cognitive processes unrelated to auditory input lead to increased activation during silence. Later, they propose that possible fluctuation in this type of basal activity in the absence of an external stimulus may play a role in auditory hallucinations.

Methods
The experiment consisted of 12 male volunteers whose brain activity was measured during silence and during an auditory stimulus ( a recording of a voice saying simple phrases) using fMRI technology. This measures oxygenated blood to certain regions of the brain and therefore activity. The subjects functional brain volume was measured during stimulus and no-stimulus sessions. Functional brain volume is the total volume of the brain activated. From this data they use statistical analysis to determine activity of background.
First they define which regions are speech-sensitive in these subjects. They compare the silent versus speech data and find the areas activated significantly during an external stimulus. This then is their speech-sensitive region they follow for the rest of the paper. It is within the auditory cortex as previously described.
Next, they want to discern the baseline activity from activation. They do this by finding the mean activation of defined voxels (basically a cubic pixel to define small regions of the brain by volume) over the silent period. Then they find time points where particular voxels are activated more than 2 SD over the mean value. From this they find the time point with the greatest amount of voxels 2 SD above the mean and where more than 2.5 % of the total voxels are activated and define this as spontaneous activation. They also define time points where the voxels are activated greater than 2SD above the mean and were common between the higher activation time points to be spontaneous activation. An example is seen in figure 1. Here they show an example of many voxels being activated significantly over what would be activated randomly or based on chance. They use a lot of statistical analysis that I am going to brush over since I am not an expert on it. They then look at the difference between activation and baseline activity through the whole brain at these defined time points. For the second level image analysis, they basically merge the results for left and right hemisphere from all the subjects. Then they remove individual variance because they are looking for something in common that is activated in all cases. This is an important bias.

Results
All of the subjects that they tested showed significantly significant spontaneous activation in both hemispheres of the brain. They used a few different statistical methods to see if the temporal activation followed some type of known function but they couldn’t find anything significant. From this, they are becoming convinced that this is just not some type of noise that is happening. The region in the left hemisphere that they found to be activated is the superior temporal gyrus, which is important for the perception of sound. They also found this activated when they look at the right hemisphere data. They also found activation in the Anterior Cingulate Cortex which is important in attention, error detection, response monitoring, and possibly many other functions.
To further ensure that this is real activation and not some type of background activation they repeat their analysis but this time compare areas of deactivation (Z<SD-2). They find that in the left hemisphere there are many more spontaneous activation time points than deactivation as opposed the right hemisphere where there are similar amounts. They use this to further narrow this activation to be specific to the left hemisphere. Also, they find that the amount of voxels greater than 2SD above the mean is greater in silence than with external stimulus in the left hemisphere and not in the left. This is consistent with the idea, that at least in the left hemisphere, there is a true spontaneous activation of the auditory cortex in silence. They go through a few other methods of statistical analysis to verify this data by looking at it from the view of the whole brain. All this leads to the same conclusions.

Conclusions
The investigators found that there is a significant activation of the auditory cortex and the ACC during silent periods. This activation is initiated in the left hemisphere with little to no variation between subjects(in right handed subjects). This is an important finding because it indicates that there is a basal activation that, if it became dysfunctional, could lead to auditory hallucinations without physical damage to the brain. Since both the auditory cortex and the ACC are activated in the silent spontaneous activation as well as in auditory hallucinations, this can explain why schizophrenics as well as those under hypnosis can experience auditory hallucinations. It does not require a structural difference in the brain. Also, this paper shows that the brain can have its own internal spontaneous activation that is not just random noise. This is intriguing in itself because it says that the brain is activated by more than just external inputs.

Paper 2 to follow. Sorry it is so late but I've been really busy at work.

 

[detta]

“Where the Imaginal Appears Real: A Positron Emission Tomography Study of Auditory Hallucinations.”
Szechtman, Woody, Bowers & Nahmias ; PNAS 1998

The authors on this paper were looking for where in the brain auditory hallucinations are processed to find out why the brain detects them as an external stimulus instead of an internal stimulus. It is a study into how our brains distinguish external auditory input from internal and how this is altered in the hallucination.

Methods
The brain scanner of choice here is the PET scan. Like an fMRI, it measures oxygenated blood in certain areas of the brain as a marker of activity but instead of using the properties of the oxygenated blood itself, it uses radioactive markers in the blood to detect where the oxygenated blood is being directed. The 8 main subjects were all males that were highly hypnotizable and able to have auditory hallucinations while hypnotized. This was all tested for before the experiments. Other subjects they used were those who were hypnotizable but could not have auditory hallucinations. The subjects were monitored under four conditions: thinking of nothing, listening to a tape after being told to listen, imagining they were listening to the tape as vividly possible, and listening to nothing while being told to listen to the tape (the hallucinatory condition). The subjects than rated the clarity of the sounds they hear as well as if they thought it was external or internal.

Results
Figure 1.
Figure 1 a shows the region of the brain activated in the hypnotizable, hallucinating subjects during hearing and hallucinating normalized to the areas activated during imagining and baseline (thinking of nothing). This area is common to both activities and not during imagining or baseline activity. This area of activation is in the ACC. Figure 1b is just a graphical representation to show that baseline and imagining conditions have similar activation while hallucinating and hearing have a similar increase. Figure 1c shows the correlation between activation of the ACC area and the subjects’ responses. The correlation between whether the subject thought the stimulus was external and the activation of the ACC region has a very high correlation coefficient, r= .95, which is a very good correlation. The correlation to clarity and activation is also very linear. This is a pretty good indicator that the increased activity in the ACC is correlated with hallucinations.
The investigators then tested the non-hallucinators and looked at the activation in their brains. Interestingly, they did not have activation in this region which strongly favors the idea that ACC is important for the creation of the hallucination. Instead, the non-hallucinators had activation in the auditory association cortex in both the hearing and the hallucinatory conditions. This is a similar region of activation as in the Hunter paper. They show this region in figure 2.

In Figure 3, they compare hallucinating versus hearing and hallucinating versus baseline for both subject groups. They noticed that not only were the regions activated in the hallucinating group much larger than the non-hallucinators but also that they also had different activation during hearing alone. While both have the auditory cortex activated, the hallucinators have a larger area activated and some of the activation is in the ACC. The authors indicate that this could mean that people who are able to have hallucinations under hypnosis are processing aural information differently than those who do not have these types of hallucinations.
From this data the authors hypothesize that since the ACC is involved with the attention system and since they instruct their subjects to pay attention to an external stimulus (They are told to listen to the tape but then none is played.), that those able to have hallucinations directed the internal event of hearing the tape to an external one. In other words, since they already had their attention directed to an external stimulus, their brains interpreted what they heard as being external.



Both these papers bring strong evidence of how auditory hallucinations are formed. The Szechtman paper shows that subjects that have hallucinations have different activation in their brains than those who do not normally have hallucinations. This activation occurs during hearing and hallucination and includes the ACC. This is important because it provides a link between just activation of the auditory cortex, which also occurs in the non-hallucinating subjects, and the ability to hallucinate. Interestingly, in periods of silence, the Hunter et al show that both the auditory cortex and the ACC can be slightly activated above baseline. The area of activation is actually reduced during hearing compared to silence. From this it seems like there is a regular system in the brain to activate these regions with or without stimulus but it does not lead to hallucinations. External stimulus seems to regulate this activation. But in those susceptible to hallucinations, there is already a baseline difference in hearing external stimulus which increases the activation in the ACC and the auditory cortex. From this, we can speculate that the differences in activation of these regions are leading to hallucinations. I find this particularly interesting because it implies that if one can manipulate these activations (i.e., drugs, stress, hypnosis), anyone could have auditory hallucinations.

 

[CosminaPrisma]

We've been neglecting the journal club...sorry about that! Have started to read what you posted. Thanks for taking the time out of your busy schedule to post that...

 

[Moonbear]

We've been neglecting the journal club...sorry about that! Have started to read what you posted. Thanks for taking the time out of your busy schedule to post that...

I'm guilty of that too. Real life (and a real life journal club) have gotten in the way. I need to catch up on what everyone has been presenting!