2021 SfNIRS Virtual Meeting Q&A: Oral session – topic 1 (fNIRS in Neurodevelopment Research)

You must be logged in to view this content.

37 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments

ReRebecca
ReRebecca (@rerebecca)
7 days ago
Prenatal exposure to an unfamiliar language: A facilitation or competition effect for the newborn’s brain linguistic processing?" Read more »

trial

Ilias Tachtsidis
Ilias Tachtsidis (@i-tachtsidisucl-ac-uk)
7 days ago
Prenatal exposure to an unfamiliar language: A facilitation or competition effect for the newborn’s brain linguistic processing?" Read more »

Hi Laura, have you also analysed the deoxyHb? If not why?

Laura Caron-Desrochers
Laura Caron-Desrochers (@lcarondesrochers)
Reply to  Ilias Tachtsidis
7 days ago

Hi Ilias,
Yes I did analyse the deoxyHb (independently of HbO), due to time constraints I did not report it here.
To sum up the deoxyHb results:
There are several channels in which we do see a significant decrease of deoxyHb in frontal, temporal and temporo-parietal regions [with FDR correction] (most overlap with channels with a significance oxyHb increase, but a few channels are significant only in one of the two Hb)
Regarding the effect of prenatal exposure: nothing is signficant (no main effect or interaction), but we did observe a marginal effect regarding group, language condition or hemisphere [with the same left-hemispheric dominance for the native language across all participants]

Thus, results from oxyHb and deoxyHb tend to overlay, but we definitly have less statistical power for deoxyHb. And this is considering that each newborn had on average 10+-3 trials per condition (resulting in a relatively good signal-to-noise ratio when averaging trials together, compared to other infant studies)

(I am definitily curious to know if there are ways to shape analyses to consider both oxyHb and deoxyHb and their relationship)

Thanks for your question!
Laura

Ilias Tachtsidis
Ilias Tachtsidis (@i-tachtsidisucl-ac-uk)
Reply to  Laura Caron-Desrochers
7 days ago

Do have a look at our poster 21st of October for some ideas: 184 Uzair Hakim, Paola Pinti, J.Adam Noah, Xiang Zhang, Paul Burgess, Antonia Hamilton, Joy Hirsch and Ilias Tachtsidis Investigation of fNIRS signals ability to infer brain activity and development of the Haemodynamic Phase Correlation signal Assembly I; TOPIC 5

Ilias Tachtsidis
Ilias Tachtsidis (@i-tachtsidisucl-ac-uk)
7 days ago
Prediction of tactile stimuli in the premature newborn brain" Read more »

How you think your BFI responses are related or will be related to oxy- and deoxy- Hb response? and is this localised? Have you measured at a different no activated region?

Victoria Dumont
Victoria Dumont (@dumontv)
Reply to  Ilias Tachtsidis
6 days ago

Hello Ilias,
BFi was shown to present a quick increase during stimulation and quick return to baseline after stimulation ends, slighter faster than HbO changes and with bigger normalized values. See Roche-Labarbe et al. 2014 for precise graphs of how these changes relate to HbO and Hb in preterm newborns during tactile stimulation.
Since we had only 1 channel we could not compare with a non-activated region in this study, but our hypotheses are in fact independent from localization. If we showed predictive activity in another region such as the frontal area, we would still conclude that preterm neonates are anticipating stimuli. In future studies, we will use multiple imaging techniques to specifically address localization questions.

Hope it answers your question, Victoria

Ilias Tachtsidis
Ilias Tachtsidis (@i-tachtsidisucl-ac-uk)
Reply to  Victoria Dumont
6 days ago

Thanks for the reply Victoria. Might worth even with one channel to create two groups of participants, one can be the control group were you place the DCS in a control non-activating region of the head. Just a suggestion.

mari
mari (@mari)
Admin
7 days ago
Prediction of tactile stimuli in the premature newborn brain" Read more »

great talk! without short separation regression can some of the difference between fixed and jitter may be due to systemic physiology?

Victoria Dumont
Victoria Dumont (@dumontv)
Reply to  mari
7 days ago

Hello Maria-Angela, thank you. We did not notice any difference in vigilance states or monitoring values between the two groups during measurement. We are analyzing monitoring parameters (heart and respiratory rates, and saturation) more closely but so far no difference arise in systemic physiology between groups. Hope it answers your question, Victoria

John Sunwoo
John Sunwoo (@bistdude)
7 days ago
Prediction of tactile stimuli in the premature newborn brain" Read more »

Thank you for your talk! My question is about the timing of the idle period after the 3 sec stimuli. What if the response would develop over the next 20-30 seconds after the stimuli end? In other words, how do we make sure that the next stimulus would not be introduced on top of the (possibly) ongoing previous response?

Last edited 7 days ago by John Sunwoo
Victoria Dumont
Victoria Dumont (@dumontv)
Reply to  John Sunwoo
7 days ago

Hello John, thank you for your question. Pre-tests and other studies on this population using tactile stimulation show that there is a quick return to baseline after stimulation ends. There could be some overlap but not 20 seconds, however, in the jittered condition when interstimulus intervals were 7 seconds, baseline stayed stable. Hope it answers your question, Victoria

Lloyd-FoxSarah
Lloyd-FoxSarah (@lloyd-foxsarah)
7 days ago
Prenatal exposure to an unfamiliar language: A facilitation or competition effect for the newborn’s brain linguistic processing?" Read more »

Great talk Laura. I wonder if you could tell me how long the sound was played during the stimulation at 35 weeks, I did not hear? And with this in mind, how short or long is sufficient stimulation (i.e. to a language) for a differential brain response to be evident five weeks later when they are born? Would you predict the same for other auditory stimuli, or is this specific to language/human sounds?

Laura Caron-Desrochers
Laura Caron-Desrochers (@lcarondesrochers)
Reply to  Lloyd-FoxSarah
7 days ago

Hi Sarah!
In each session, fetuses were exposed to a 5-min story repeated twice alternating language versions. (i.e. a total of 10 min in French, and 10 min in the foreign language).
The prenatal exposure sessions started from the 35th gestational week up to birth, and mothers were instructed to do a minimum of one session every two days, ideally everyday (On average mothers completed 25 days of prenatal exposure).

Regarding how short or long is sufficient stimulation… it is actually a great question that I don’t have the answer for now. We explored this question by doing linear multiple regressions using the amount of prenatal exposure as predictors, and what I can say for now is that there is not a linear relationship with Hb response/Hb differences across hemispheres or conditions. My hypothese is that we might have unsufficient variability in the amount of prenatal exposure + not many participants within each of our experimental prenatal groups. Due to the covid, we were not able to go on the postnatal ward and had to interrupt our testings (still applying now). But there are interesting EEG studies from Partanen and colleagues that linked the amount of prenatal expsure with the mismatch response at birth regarding speech phonemes.

And finally, even if our data cannot answer your last question, I do think that other complex auditory stimuli exposed during gestation might shape brain processing and that we could see a “prenatal effect” (there are a few studies that looked at musical melodies and EEG response or somatocardiac modulation). But I still think that the brain is tuned to speech (preferential / more refined processing of acoustical parameters and dynamics in speech and human sounds), which should facilitates learning!

Thanks for your questions!
Laura

Lloyd-FoxSarah
Lloyd-FoxSarah (@lloyd-foxsarah)
7 days ago
Prediction of tactile stimuli in the premature newborn brain" Read more »

Lovely talk Victoria, really clear, though a lot to take in during 8 mins, you did well! I wanted to ask two Qs, regarding the repetition supression pattern and ommision response. 1., would you predict the same pattern if you had done this 1-2 weeks after they were born if at term, or would you predict any differences due to prematurity? and 2. have you related any of your work to foetal studies using ultrasound, which usually are based on auditory stimuli, but that have looked at habituation of behavioural response to stimulation?

Victoria Dumont
Victoria Dumont (@dumontv)
Reply to  Lloyd-FoxSarah
6 days ago

Thank you Sarah,
We do not necessarily expect a big difference based on gestational age, but one based on postnatal experience. We did not have time to present all the data but we also measured our subjects at 35 weeks corrected (2 weeks of postnatal NICU experience) and the results show that repetition suppression is altered in Fixed condition, whereas prediction and RS in the Jittered condition are preserved. As RS was observed in the Jittered condition, it suggests that NICU stay may not be the reason since both groups were similar in that regards. Instead, it could reflect a developmental shift from a wide sensitivity at birth to both deterministic and probabilistic events, to a specific attraction to probabilistic events two weeks later. Literature reports that infant’s preference in 7 and 8 months old (see Kidd et al., 2012) is drawn toward moderately predictable events instead of fully predictable or unpredictable events. Thus, maybe two to three weeks of extrauterine environment could be enough for the baby’s brain to specialize in the more relevant probabilistic events ?
In any case, that would be interested to compare with term-neonates.
Foetal studies of RS show quick RS/habituation in foetuses with auditory stimuli, but the temporal structure is highly predictable with fixed, short ISI. Our results at 33 weeks are consistent with that, in a different modality and a mush longer time frame, but the most interesting thing is the comparison with Jittered ISI, which we haven’t seen in the foetus literature, and with 35 weekers which is of course specific to the preterm population.

Hope it answers your question, Victoria

Sara Sanchez-Alonso
Sara Sanchez-Alonso (@sara-sanchez-alonso)
7 days ago
Hemodynamic Correlates of Speech Processing Using fNIRS: Assessing the Effect of Bilingualism in Young Infants" Read more »

Hi Borja, very nice talk! I was wondering why you think you didn’t see differences in backward versus forward speech relative to prior studies that did find differences?

BlancoBorja
BlancoBorja (@blancoborja)
Reply to  Sara Sanchez-Alonso
7 days ago

Thank you Sarah,

We believe there might be several explanations:
1) infants tested during natural sleep (although previous works showing differences were also conducted in infants during natural sleep)
2) differences in preprocessing (we include a step for reducing physiological confounds), population (newborns, 4-month-olds) and data quality
3) similar auditory properties (or not salient enough differences) between forward and backward stimuli.

I would love to hear your thoughts, we can discuss more these results in GatherTown. I can also show you examples of our forward and backward speech stimuli.

Borja

Sara Sanchez-Alonso
Sara Sanchez-Alonso (@sara-sanchez-alonso)
Reply to  BlancoBorja
5 days ago

Thank you, Borja. I would be interested in knowing more about your step for reducing physiological confounds in infants. Unfortunately, I could not attend the Monday social session, but will attend on Thursday. Also, I was looking at your poster on precision functional mapping using HD-DOT, do you have a pre-print ready? Very interesting poster!

Last edited 5 days ago by Sara Sanchez-Alonso
Lauren Emberson
Lauren Emberson (@laurenemberson)
7 days ago
Prediction of tactile stimuli in the premature newborn brain" Read more »

Amazing talk. Super super interesting results and a lot to think about. My main question is similar to Sarah’s: how are you thinking about these findings compared to the ones we found in preterm infants later in their life? do you think that the differences are in the modalities and task or do you think that if you tested tactile prediction at 6 months corrected you would find the same impairments that we found in AV prediction? I know you cannot answer this with your data but I would love to hear about your thinking. thanks.

Victoria Dumont
Victoria Dumont (@dumontv)
Reply to  Lauren Emberson
6 days ago

Thank you Lauren,
We think that we would likely find similar deficits to what you found at 6 months, because it is not only the untimely birth but also the weeks of NICU stay with multiple painful procedures and inadequate sensory inputs that build upon preterm birth, to generate atypical sensory profiles. It seems likely that this happens in all sensory modalities. However, we believe that tactile processing is essential to early development and first social interactions, so maybe it is more robust than later, less essential modalities ? or maybe visual networks are more impacted by preterm birth ? As a first element of response, we did not have time to present all the data but we also measured our subjects at 35 weeks corrected (2 weeks of postnatal NICU experience) and the results show that repetition suppression is altered in the fixed condition, whereas prediction in the Jittered condition is somewhat preserved. We plan on following up on that question with older preemies to see how it evolves.
Hope it answers your question, Victoria

Ilias Tachtsidis
Ilias Tachtsidis (@i-tachtsidisucl-ac-uk)
7 days ago
Increased task-based connectivity across development in a social cognition task: Links to developmental outcomes?" Read more »

Have you look at the deoxyHb connectivity?

Sabrina Burr
Sabrina Burr (@smburr)
Reply to  Ilias Tachtsidis
7 days ago

Yes, we looked at both oxy and deoxy and we found that there was some increase in deoxy connectivity from 6 to 24 months but it was only unilateral. The most noticeable increases occurred in oxy where we found greater connectivity both within and between hemispheres 

Ilias Tachtsidis
Ilias Tachtsidis (@i-tachtsidisucl-ac-uk)
Reply to  Sabrina Burr
6 days ago

Thanks Sabrina. I am wondering if the deoxyHb connectivity maps represent a different brain physiology from the oxyHb connective maps in infants. Something to think about. Again great talk.

Caroline Kelsey
Caroline Kelsey (@caroline-kelseychildrens-harvard-edu)
7 days ago
Increased task-based connectivity across development in a social cognition task: Links to developmental outcomes?" Read more »

Hi Sabrina,

Wonderful talk! I am excited to see the task-based connectivity analyses being used with the BEAN data. I am new to this technique and was curious to learn more about some of the processing decisions. First, I was wondering if you had any thoughts or suggestions as to what type of event-based paradigms these analyses would work well on and if there were suggestions for which time period to use (e.g., longer vs shorter stimuli; a minimum amount of time; selecting the time only during the stimuli presentation or also using the ISI/baseline period as well, etc.)
For my own edification, I was also interested to know why the FIR filter was used and how the analyses would change if one used a butterworth filter and a non-glm approach.
Did you use global signal regression as part of your processing pipeline and why or why not?
Finally, I was curious to hear how you are thinking about the null results for the brain-behavior associations. Do you think if the task was something more closely related to the mullen outcomes (e.g., language, etc.) that perhaps you would see an association?

Thank you again for this important work!

Lauren Emberson
Lauren Emberson (@laurenemberson)
7 days ago
Attention in Toddlers is Related to Individual Differences in Executive Functioning and Parental Risk for Attention Deficits" Read more »

Anastasia, thanks for this great talk! This is very interesting work. I think I figured this out through the talk but the maternal self-report, is that of the toddler’s/children’s symptoms or of their own?

Luca Pollonini
Luca Pollonini (@lpolloni)
Editor
7 days ago
Links Between Socio-Economic Disadvantage, Neural Function, and Working Memory in Early Childhood" Read more »

Great job Xinge!

BoasDavid
BoasDavid (@boasdavid)
Reply to  Luca Pollonini
7 days ago

I agree. Great to see you Xinge

John Sunwoo
John Sunwoo (@bistdude)
7 days ago
Links Between Socio-Economic Disadvantage, Neural Function, and Working Memory in Early Childhood" Read more »

Thanks for the interesting talk – I might have missed it but was there any difference in the task acurracy in lower SE status? You showed us that the activations (Lateral PFC) were lower in lower SE status, but what if the lower activation was due to short in their attention abilities, which may be different and accounted from impeding cognitive ability in their development?

Lauren Emberson
Lauren Emberson (@laurenemberson)
7 days ago
Preschoolers’ prefrontal cortex activation in action sequence planning in a naturalistic situation" Read more »

Hi Lisanne– thanks for your awesome talk and this really important new approach to fNIRS with toddlers. I would love to hear more about how you are thinking that this naturalistic setting is helping with fNIRS recordings. Do you think that having the toddlers being freely moving and doing these kind of naturalistic tasks is helpful in getting more compliance? If so, which of these do you think is most important? The natural tasks (e.g., really playing with duplo) or being able to move? This is a very exciting direction of research!

Lisanne Schroer
Lisanne Schroer (@lisanneschroer)
Reply to  Lauren Emberson
7 days ago

Hi Lauren,
Thanks a lot!

We believe that an engaging task, such as this naturalistic study with Duplo is the most important. A previous study by Paola in our lab consisted of similar equipment with toddlers in a conventional experimental sitting in front of a screen watching videos. In this study, children get bored after a little bit, as these tasks are not very engaging for this age range.

In our current study, the naturalistic task definitely improved compliance. Children are very excited, and don’t even need to be instructed to sit still. They do not complain at all, and none dropped out of the task. As long as children are interested in what they do, they are very compliant.

I would say that for some children it does help that they can move for their compliance, but generally the most important is that the task is fun and engaging. Luckily, naturalistic tasks are more often fun for this age range compared to the standardized tasks.

Lisanne

Ilias Tachtsidis
Ilias Tachtsidis (@i-tachtsidisucl-ac-uk)
7 days ago
Neural responses to touch in naturalistic mother-infant interactions" Read more »

Trinh i agree with you that we need to understand better the haemodynamic response in infants do also check out our recent review paper https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6176492/

Trinh Nguyen
Trinh Nguyen (@trinh-nguyen)
Reply to  Ilias Tachtsidis
7 days ago

Thank you, Ilias! I’ll definitely have a look at the paper.

Sara Sanchez-Alonso
Sara Sanchez-Alonso (@sara-sanchez-alonso)
7 days ago
Association of psychosocial adversity and social information processing in urban Bangladeshi children: a fNIRS study" Read more »

Such a great talk, Laura! I found your connections between BRIGHT and BEAN particularly useful. Are the differences between the auditory and visual domain also observed in the BRIGHT data?

Laura Pirazzoli
Laura Pirazzoli (@laurapirazzoli)
Reply to  Sara Sanchez-Alonso
5 days ago

Hi Sara,

if you refer to the associations with adversity, in BEAN we mostly found them with discrimination in the auditory domain while discrimination in the visual domain seems to be spared by psychosocial adversity. In BRIGHT it looks like both domains are impacted by maternal depression. As far as the group results for social visual and auditory discrimination are concerned we also find very different patterns. I think we need to think carefully about all the factors on which these two samples differ, to understand what might explain these differences.

Last edited 5 days ago by Laura Pirazzoli
Sara Sanchez-Alonso
Sara Sanchez-Alonso (@sara-sanchez-alonso)
Reply to  Laura Pirazzoli
5 days ago

Thanks, Laura. Looking forward to seeing the publication of these findings.

John Sunwoo
John Sunwoo (@bistdude)
7 days ago
Neural responses to touch in naturalistic mother-infant interactions" Read more »

Thanks for the talk! I am wondering how the Oxygenation (=HbO-HbR) response was different in your study? As you showed, HbR did not show the same results and I thought that Oxygenation can sort out false activation (such as increases in both HbO and HbR, and vice versa).

Trinh Nguyen
Trinh Nguyen (@trinh-nguyen)
Reply to  John Sunwoo
6 days ago

Hi John, so our HbR values show decreases in both same (IFG) and different (mPFC) channels than where HbO values are significantly increased. We can take the channels with “concordant” responses (increase in HbO, decrease in HbR) as robust responses, but as Ilias pointed out, in infants it’s not exactly clear that such an inverse relation is as valid as in adults. We’ll therefore probably need to have a closer look at the data.