A Neural Response to “Trigger” Stimuli in PTSD?

By Neuroskeptic | January 17, 2016 8:04 am

A new paper in the prestigious Journal of Neuroscience makes some exciting claims about the neurobiology of PTSD – but are the methods solid?

Canadian researchers Mišić et al. used magnetoencephalography (MEG) to measure neural activity in four groups: traumatized Canadian soldiers, non-traumatized soldiers, civilians with mild traumatic brain injury, and healthy civilians. They found that

Soldiers with PTSD display inter-regional hypersynchrony at high frequencies (80–150 Hz), as well as a concomitant decrease in signal variability. The two patterns are spatially correlated and most pronounced in a left temporal subnetwork, including the hippocampus and amygdala. We hypothesize that the observed hypersynchrony may effectively constrain the expression of local dynamics, resulting in… functional networks becoming “stuck” in configurations reflecting memories, emotions, and thoughts originating from the traumatizing experience.

Here’s some of the results. In this image, we see the effect of exposure to “triggering” war-related stimuli on the soldiers’ brains. The spiderweb of red lines shows increased synchrony across the brain, especially in the high frequency gamma band (both low and high gamma), across both military groups (PTSD and non-PTSD).

ptsd_misic

Mišić et al. perform various group comparisons (not pictured here) to conclude that brain activity in response to triggering stimuli shows hypersynchrony and reduced signal variability – although the soldiers with PTSD also show a smaller change in brain activity following the triggers, suggesting that they may have been, so to speak, ‘triggered’ at baseline.

Now that’s rather interesting. But we should be cautious. MEG doesn’t just detect neural activity. It also, unfortunately, picks up artifacts, magnetic fields generated outside the brain. These include ocular artifacts, from movement of the eyes, and muscle artifacts, that originate from the activity of various muscles around the head, face, and neck.

Worryingly, these non-neural artifacts are notorious for generating high frequency noise that contaminates the gamma band – which is just the frequency band where Mišić et al. found their key results.

So how did Mišić et al. mitigate this problem? After noting that they used a beamformer approach to data analysis, they say that

MEG beamformers have been shown to be effective at suppressing ocular and nonocular artifacts, including cardiac and muscle activity (Muthukumaraswamy, 2013), obviating further artifact correction.

That’s all they have to say about artifacts. It seems a little blasé to me, but don’t take my word for it: here’s what Suresh Muthukumaraswamy has to say, in the only paper that Mišić et al. cite as support (emphasis mine).

While beamformer source-reconstruction images are not explicitly an artifact-removal algorithm, high-frequency artifacts in these images tend to localize to their source locations, making them relatively easy to spot in source-reconstruction images. Spatially filtered, source-space “virtual” sensors can then be subjected to subsequent analyses.

However… for beamformer virtual electrodes, caution must still be exercised because the channels are not necessarily artifact-free, particularly if the spatial filters are relatively coarse and the artifacts are relatively large.

Muthukumaraswamy went on to state a number of “Recommendations for the Collection, Analysis, and Presentation of High-Frequency MEG/EEG Experiments”. Mišić et al. seem to fall short of many of these. For instance

#1. Presentations of data that use statistical analysis only, without first presenting spectral or spatial representations should be avoided.

Mišić et al. only present statistical analyses, not raw data.

#2. Presentation of (time-)frequency spectra is critically important. The spectrum should be presented in a way such that the full bandwidth of the high-frequency activity of interest is visible.

There are no time-frequency spectra in the paper.

#3. Presentation of spatial maps (topographic maps and/or source localizations) for high-frequency activity is important. When broadband activity arises near the edge of the sensor/electrode montage, or the source solution space, this may be an indicator of electromyographic [muscle] contamination

No spatial maps of high frequency activity are provided. Mišić et al. show spatial maps of synchrony and variability, which are statistically derived from the activity, but they never plot the activity itself.

#9. Collection of EOG [electrooculogram; i.e. to enable detection of ocular artifacts] is highly desirable for both MEG and EEG. Where feasible and appropriate, eye-tracking should also be considered. Again, this is particularly important for frontal and temporal sources.

No EOG or eye-tracking was collected in Mišić et al. so the role of eye movements is unknown. Their key results come from frontal and temporal sources.

In his paper Muthukumaraswamy gives an example of how insidious non-neural artifacts can be, pointing to a 2011 MEG study from his own lab which seemed to show high gamma activity in the cerebellum associated with making a response to a stimulus using a joystick. But this was, Muthukumaraswamy says, almost certainly nothing more than muscles in the neck which contracted as people leaned forward to grasp the joystick.

Here’s Figure 2 from Muthukumaraswamy’s paper, showing the issue. In panel C, the time-frequency plot, we see that the stimulus-evoked activity is “broad-band”, spanning the range of frequencies from 50 to 150 Hz. No neural signal has such as wide range so Muthukumaraswamy says that his data was “highly likely” to be an artifact.

Muthukumaraswamy_megIt’s easy to imagine how Mišić et al.’s traumatized soldiers might have reacted more strongly to the triggering stimuli in terms of muscle movements: maybe they showed jaw clenching, tenseness, postural changes, different eye movements, etc.

Whether Mišić et al.’s data contain such artifacts or whether they explain these results is impossible to determine: I don’t know if muscle or ocular artifacts could increase synchrony across brain regions, as Mišić et al. found. But I would have expected peer reviewers (at the Journal of Neuroscience no less) to have requested some checks and balances to clarify this point.

Edit 19th January 2016: I’ve clarified the discussion of Mišić et al.’s group comparisons and the passage on Muthukumaraswamy’s figure.

ResearchBlogging.orgMišić B, Dunkley BT, Sedge PA, Da Costa L, Fatima Z, Berman MG, Doesburg SM, McIntosh AR, Grodecki R, Jetly R, Pang EW, & Taylor MJ (2016). Post-Traumatic Stress Constrains the Dynamic Repertoire of Neural Activity. The Journal of Neuroscience, 36 (2), 419-31 PMID: 26758834

ADVERTISEMENT
  • Aina Puce

    Interesting questions you have raised about this study and others like it. See also the paper below for a nice practical example of the visualization and removal of artifacts in source vs sensor space.

    PMID: 23847508

    Hipp JF, Siegel M. (2013) Dissociating neuronal gamma-band activity from cranial and ocular muscle activity in EEG. 23847508 Front Hum Neurosci. 2013 Jul 10;7:338. doi: 10.3389/fnhum.2013.00338. eCollection 2013.

  • feloniousgrammar

    After the silent MRI is invented and ubiquitous we can perhaps see if there are broad differences between images from current machines vs. those tested with the hypothetical silent machine to see if there are more signals in more places in a human brain being assaulted with a bizarre series of loud and overbearing noises in changing patterns that have nothing to do with the test. I’d go as far as to submit that being crammed in a culturally blind narrow tube has an impact and is directly related to the timing and quality of the thoughts and the associations that the experiment triggered.

  • JR KING

    > “The stimulus-evoked activity is “broad-band”, spanning the range of
    frequencies from 50 to 150 Hz. No neural signal has such as wide range.”

    I don’t think that this is quite correct. Here’s an example from a colleague of mine: http://www.unicog.org/publications/Cereb.%20Cortex-2014-El%20Karoui-cercor-bhu143.pdf.

    In MEG, my experience is that although high gamma can be difficult to observe, it’s not impossible. One also has to bear in mind that inter individual variability could increase the variance of the gamma range.

    That being said, I haven’t seen in my data such late and prolonged broad band effect with EEG or MEG (but I have not used their protocol either).

    Another concern is that they find prominent effects in the subcortical areas. By definition these distant regions are difficult to record from with MEG, and have the tendency to project onto a large number of sensors … a property shared with muscular artifacts.

    Certainly more data viz and control analyses could strengthen their claims. But JNeuro doesn’t accept supplementary materials.

  • Aina Puce

    I have been thinking a lot about this post because the issue you raise is such a hard one to nail down analytically for everyone right now.

    One thing to think about is that it is hard to remove an artifact from a signal if that artifact has itself not been properly characterized. To that end, one way to potentially begin to attack the EMG issue in MEG/EEG studies might be to greatly increase the sampling rate and the bandpass filtering of MEG/EEG data at acquisition time, so that muscle artifacts might be more clearly separated from neural activity in subsequent analyses. We have been experimenting with increasing our bandpass and sampling rates in the lab as we experiment with ICA methods also, and have been thinking about the differentiation of artifacts in sensor/source space. One argument for increasing sampling rates/filter bandpass comes from the success of removing MRI scanning artifacts in simultaneously acquired EEG recordings – the compound signal is sampled at high rates, so that the MRI scanning artifact can be properly characterized and then removed. Having said that, the MRI scanning artifact is very stereotypical and lends itself to removal with different methods, including ICA. EMG activity is not stereotypical – other than showing up in typical locations on the head and having the higher frequency activity in the MEG/EEG signal it can occur with variable persistence and at different times during the experiment. This makes it so difficult to remove. And as you point out, that activity can correlate with the task if the task is difficult or distressing.

    So perhaps we could begin to fare better at removing EMG activity from MEG/EEG recordings if the compound signal was sampled better? Whether we can successfully remove the EMG artifact after changing our data acquisition remains to be determined.

  • Bratislav Misic

    Dear Neuroskeptic,

    thanks for your interest in our recent paper.

    The issue that you have raised – whether gamma-band, source-space synchrony could be artefactual – is a valid criticism and one that cannot be definitively ruled out because we did not record EOG or eye movements. The fact that saccadic eye movements can generate band-limited gamma-band oscillations has been well documented in recent years (e.g. Carl et al., 2012; Hipp and Siegel, 2013; see comment by Aina below). This certainly should have been mentioned as one of the limitations of the study.

    At the same time, there are several reasons to be skeptical about the role of ocular and other artefacts in the present results:

    1) While saccades and microsaccades have been shown to induce local gamma rhythms, we studied phase synchrony between sources. Phase synchrony between remote sites is, in theory, independent of local oscillatory power at those sites. Thus, condition- or group-driven changes in phase synchrony do not necessarily have to accompany changes in local power.

    2) Any artefactual interference due to eye movements would simultaneously/instantaneously project to all sources, manifesting as zero-lag phase synchrony. To avoid this potential confound, we estimated phase synchrony in terms of the weighted phase lag index (wPLI; Vinck et al., 2011), a conservative measure that attenuates these spurious 0- and 180-degree phase relationships.

    3) While it would be impossible to do a time-frequency analysis (these are resting state recordings), we did perform a frequency (spectral) analysis (Figure 6), as suggested by one of our reviewers. We found evidence for group differences (Figure 6), but did not find evidence for condition differences, such as those shown in Figure 1. The group spectral power analysis showed many parallels with the group synchrony analysis, but many of these similarities were also observed outside of the ‘low’ and ‘high’ gamma band (e.g. reduced alpha-band synchrony and power). Conversely, there were also several differences between the two analyses. For instance, while the hypersynchrony results were more pronounced in the high gamma band (80-130 Hz), the spectral power results were more pronounced in low gamma (30-80 Hz). The implications of these results are discussed in greater detail in the Discussion (p. 428).

    4) On a related note, a recent paper by Ming-Xiong Huang and colleagues at UCSD also found evidence of increased gamma-band activity in PTSD even after removal of ocular and muscular artefacts using ICA (Huang et al., 2014). This suggests that gamma-band activity in this population is at least partly non-artefactual.

    5) Many of the regions implicated in the present analysis (hippocampus, amygdala, superior temporal sulcus, etc.) are consistently implicated in PTSD (Bremner, 2006; Pitman et al., 2012). The importance of these structures in PTSD has been demonstrated not just in MEG (e.g. Engdahl et al., 2010; Georgopoulos et al., 2010; Huang et al., 2014), but also in structural MRI (grey matter volume; Kitayama et al., 2005), fMRI (e.g. Etkin and Wager, 2007) and PET (e.g. Bremner et al., 2003). Most importantly, hyperconnectivity has also been reported in PTSD using other modalities (e.g. Brown et al., 2014). Thus, it is not unexpected that we would observe most of the effects in these areas, and the prominence of medial temporal and frontal areas does not necessarily imply contamination from other sources.

    6) As Neuroskeptic points out, a key result concerns the synchrony differences between pre- and post-trigger recordings. If artefacts were present in one group (PTSD) in a single condition (post-trigger resting state), and were phase locked and time-locked, these could potentially dominate all subsequent analyses, which is not the case in the present data. For instance, Figure 2 captures a group effect, differentiating the two military populations from the two civilian populations. If the post-trigger resting state in PTSD were dominated by artefacts, we might expect to see a dominant contrast that differentiates that condition from all others.

    We would also like to offer a couple of further comments regarding the critique itself:

    1) The results of the study were misrepresented in the post. Namely, the contrast in Figure 1 (a part of which is shown as the first figure in the blog post) does not suggest that PTSD soldiers are more affected by the triggering procedure, but quite the opposite: they are less affected by the procedure, compared to control soldiers. The subsequent group analysis (Figure 2) shows that this may be because of elevated baseline synchrony in the PTSD soldiers.

    2) Towards the end of the post, Neuroskeptic mentions that “No neural signal has such a wide range”, implying that any experimental effect over such a broad band would necessarily be artefactual. While we agree that a task-dependent, time locked response would be unlikely to occupy such a broad band, we again remind our colleague that the present data are resting state recordings, and that the analyses describe group effects, and group x condition interactions, which may be associated with broadband changes.

    3) Muthukumaraswamy (2013) does mention the following:

    “While beamformer source-reconstruction images are not explicitly an artifact-removal algorithm, high-frequency artifacts in these images tend to localize to their source locations, making them relatively easy to spot in source-reconstruction images. Spatially filtered, source-space “virtual” sensors can then be subjected to subsequent analysis”

    Thus, artefactual power would be found close to origin of the artefact (the eyes). This is commonly observed as increased canonical power localized to the eyes when the source model is constrained to a head space. In the present study we confined the source model to the brain only.

    Further, Muthukumaraswamy (2013) also mentions:

    “#3. When broadband activity arises near the edge of the sensor/electrode montage, or the source solution space, this may be an indicator of electromyographic (muscle) contamination.”

    In the present study, most of the differences were in deep structures (e.g. hippocampus, amygdala), at a distance from the edge of the source solution space.

    The limited nature of anonymized, 2- or 3-person peer review and the typical response-counter response format often precludes a deeper discussion of all the issues that might arise in modern neuroimaging research, so we appreciate the interest and suggestions from Neuroskeptic and the other readers (@cswasserman1, @andrewthesmart, @ajshackman, @neuropathos, @aina_puce). We believe that authors should be held accountable even after a paper is published, and welcome any further comments and suggestions from the community.

    Bratislav (@misicbata), Ben (@btdunks), Zainab (@zeefatima), Marc, Sam, Randy (@ar0mcintosh), Elizabeth and Margot

    Bremner JD, Vythilingam M, Vermetten E, Southwick SM, McGlashan T, Staib LH, Soufer R and Charney DS. (2003). Neural correlates of declarative memory for emotionally valenced words in women with posttraumatic stress disorder related to early childhood sexual abuse. Biol Psychiatry, 53(10):878-889.

    Bremner JD. (2006). Traumatic stress: effects on the brain. Dialogues Clin Neurosci, 8(4):445-461.

    Brown VM, LaBar KS, Haswell CC, Gold AL, Gold AL, McCarthy G, Morey RA (2014) Altered resting-state functional connectivity of basolateral and centromedial amygdala complexes in posttraumatic stress disorder. Neuropsychopharmacology 39:351–359

    Engdahl B, Leuthold AC, Tan H-RM, Lewis SM, Winskowski AM, Dikel TN, Georgopoulos AP (2010) Post-traumatic stress disorder: a right temporal lobe syndrome? J Neural Eng 7:066005.

    Etkin A, Wager TD (2007) Functional neuroimaging of anxiety: a metaanalysis of emotional processing in PTSD, social anxiety disorder, and specific phobia. Am J Psychiat 164:1476–1488.

    Georgopoulos AP, Tan HM, Lewis S, Leuthold A, Winskowski A, Lynch J, Engdahl B (2010) The synchronous neural interactions test as a functional neuromarker for post-traumatic stress disorder (PTSD): a robust classification method based on the bootstrap. J Neural Eng 7:016011.

    Hipp JF, Siegel M. (2013) Dissociating neuronal gamma-band activity from cranial and ocular muscle activity in EEG. 23847508 Front Hum Neurosci. 2013 Jul 10;7:338. doi: 10.3389/fnhum.2013.00338.

    Huang MX, Yurgil KA, Robb A, Angeles A, Diwakar M, Risbrough VB, Nichols SL, McLay R, Theilmann RJ, Song T, Huang CW, Lee RR and Baker DG. (2014). Voxel-wise resting-state MEG source magnitude imaging study reveals neurocircuitry abnormality in active-duty service members and veterans with PTSD. NeuroImage Clin, 5, 408-419.

    Muthukumaraswamy SD (2013) High-frequency brain activity and muscle artifacts in MEG/EEG: a review and recommendations. Front Hum Neurosci 7:138

    Pitman RK, Rasmusson AM, Koenen KC, Shin LM, Orr SP, Gilbertson MW, Milad MR, Liberzon I (2012) Biological studies of post-traumatic stress disorder. Nat Rev Neurosci 13:769–787

    Vinck M, Oostenveld R, van Wingerden M, Battaglia F and Pennartz CM (2011). An improved index of phase-synchronization for electrophysiological data in the presence of volume-conduction, noise and sample-size bias. NeuroImage, 55, 1548-1565.

    • http://blogs.discovermagazine.com/neuroskeptic/ Neuroskeptic

      Many thanks for this very thoughtful and informative reply. I’ll respond in detail tomorrow!

    • http://blogs.discovermagazine.com/neuroskeptic/ Neuroskeptic

      Thanks again Bratislav for this very detailed reply. In response to your points 1-6, I don’t have much to add – these are all good points that strengthen your case, especially #2 about the phase-lag issue. That seems like a good way to exclude ocular artifacts.

      Re: the final 3 points

      1) The results of the study were misrepresented in the post. Namely, the contrast in Figure 1 (a part of which is shown as the first figure in the blog post) does not suggest that PTSD soldiers are more affected by the triggering procedure, but quite the opposite: they are less affected by the procedure, compared to control soldiers.

      You’re right, this was unclear – I have updated the post!

      2) Towards the end of the post, Neuroskeptic mentions that “No neural signal has such a wide range”, implying that any experimental effect over such a broad band would necessarily be artefactual. While we agree that a task-dependent, time locked response would be unlikely to occupy such a broad band, we again remind our colleague that the present data are resting state recordings, and that the analyses describe group effects, and group x condition interactions, which may be associated with broadband changes.

      That’s true. My “no neural signal” statement was a commentary on Muthukumaraswamy’s example from his 2013 paper (itself based on Kennedy et al. 2011). I’ve updated the text to clarify that it’s not a reference to your study.

      Regarding 3), I must confess I’ve not used beamforming myself. Does confining the source model to the brain mean that signals from outside the brain are not modelled, and hence excluded? Or might they appear as signals that appear to come from the brain regions near the true source?

      The limited nature of anonymized, 2- or 3-person peer review and the typical response-counter response format often precludes a deeper discussion of all the issues that might arise in modern neuroimaging research, so we appreciate the interest and suggestions from Neuroskeptic and the other readers (@cswasserman1, @andrewthesmart, @ajshackman, @neuropathos, @aina_puce). We believe that authors should
      be held accountable even after a paper is published, and welcome any further comments and suggestions from the community.

      I couldn’t agree more! Thanks again!

  • Sylvester Fernandes

    Can this actual phenomenon be explained by the concepts of modulomathematics?

  • Justin Hallinan

    Very interesting topic. Thank you!

  • Pingback: A Neural Response to “Trigger” Stimuli in PTSD? – usa health centar()

NEW ON DISCOVER
OPEN
CITIZEN SCIENCE
ADVERTISEMENT

Neuroskeptic

No brain. No gain.

About Neuroskeptic

Neuroskeptic is a British neuroscientist who takes a skeptical look at his own field, and beyond. His blog offers a look at the latest developments in neuroscience, psychiatry and psychology through a critical lens.

ADVERTISEMENT

See More

@Neuro_Skeptic on Twitter

ADVERTISEMENT

Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Collapse bottom bar
+