Knowledge Vault 3/63 - G.TEC BCI & Neurotechnology Spring School 2024 - Day 5
Exploration of bottlenose dolphin auditory processing
using ‘high density’ 16 channel EEG
Matt Schalles, Aurora University (USA)
<Resume Image >

Concept Graph & Resume using Claude 3 Opus | Chat GPT4 | Llama 3:

graph LR classDef hearing fill:#f9d4d4, font-weight:bold, font-size:14px; classDef echolocation fill:#d4f9d4, font-weight:bold, font-size:14px; classDef brain fill:#d4d4f9, font-weight:bold, font-size:14px; classDef research fill:#f9f9d4, font-weight:bold, font-size:14px; classDef methods fill:#f9d4f9, font-weight:bold, font-size:14px; classDef results fill:#d4f9f9, font-weight:bold, font-size:14px; A[Matt Schalles] --> B[Wide hearing range,
exceeds humans. 1] A --> C[Echolocation: clicks, echoes
for navigation. 2] B --> D[Sensitive to high
echolocation frequencies. 3] A --> E[More surface area,
folds than humans. 4] A --> F[1960s-70s invasive recordings
mapped auditory cortex. 5] A --> G[1980s non-invasive EEG:
strongest midline AEPs. 6] A --> H[Few previous studies
when lecturer joined. 7] A --> I[Ocean enclosures, underwater
sounds, head EEG. 8] A --> J['Bite plate' head
restraint during EEG. 9] A --> K[Out-of-water recordings on
mats with 'jawphone'. 10] A --> L[Gold cup electrodes,
silicone suction cups. 11] A --> M[Splash-proof amplifier housing,
panel mount connections. 12] A --> N[Long 20-30m electrode
cables to dock. 13] A --> O[Little standardization, explored
techniques empirically. 14] O --> P[Mastoid-like reference best
signal-to-noise ratio. 15] O --> Q[Hundreds to thousands
of trial repetitions. 16] O --> R[Faster rates reduced
amplitude, preserved AEPs. 17] O --> S[Habituation decreased AEP
magnitude over trials. 18] O --> T[AEP polarity reversal
matches Russian studies. 19] O --> U[Average reference cleaner
than single/filtering. 20] O --> V[Submerged recordings attenuated
vs. in-air. 21] V --> W['Shallow bite plate'
blowhole/electrodes in air. 22] A --> X[Oddball paradigm: attention,
standard/deviant/target tones. 23] X --> Y[Dolphins respond to
target, ignore others. 24] X --> Z[Larger AEPs for
deviants than standards. 25] Z --> AA[Preliminary: AEP magnitude
varies with attention. 26] A --> AB[Dolphin AEPs noisier,
techniques improve quality. 27] AB --> AC[In-air clean but
unnatural, submerged vice-versa. 28] A --> AD[Wearable EEG loggers
for natural behaviors. 29] A --> AE[Collaborative research effort
across organizations, trainers. 30] class A,H,O research; class B,C,D hearing; class E brain; class F,G,I,J,K,L,M,N,P,Q,R,S,T,U,V,W,AB,AC,AD methods; class X,Y,Z,AA results; class AE echolocation;


1.- Dolphins have a very wide hearing range up to 150 kHz, far exceeding humans, which they use for echolocation to navigate.

2.- Dolphins emit loud, broadband clicks from their forehead and hear echoes through their jaws to build an acoustic representation of their environment.

3.- Dolphin hearing is most sensitive at the high frequencies used for echolocation. Their auditory system is adapted for their underwater environment.

4.- Dolphin brains have more surface area and cortical folds than human brains. Brain anatomy and electrode placement options differ between species.

5.- Early dolphin auditory processing research in the 1960s-70s involved invasive intracranial recordings that identified auditory responses and mapped auditory cortex.

6.- The first detailed non-invasive dolphin EEG study published in the 1980s found the strongest auditory evoked potentials (AEPs) at midline electrodes.

7.- When the lecturer joined the dolphin EEG project in 2018, there were only a few previous non-invasive and invasive recording studies.

8.- The dolphins were tested in netted ocean enclosures. Sounds were played underwater while electrodes recorded EEG from the dolphin's head surface.

9.- Dolphins were positioned on a "bite plate" to keep their heads still relative to the sound source during EEG recordings.

10.- Out-of-water recordings were also done with dolphins temporarily beached on mats. Sounds were played through a "jawphone" for in-air testing.

11.- The lecturer used gold cup electrodes housed in silicone suction cups to isolate the recording sites from the surrounding saltwater.

12.- The 16-channel EEG amplifier was mounted in a splash-proof housing with panel mount electrode connections to protect it in the ocean environment.

13.- Long electrode cables of 20-30 meters were needed to connect the dolphins to recording equipment on the dock.

14.- There was little standardization of recording methods across previous dolphin EEG studies, so the lecturer explored different techniques empirically.

15.- The lecturer found that a mastoid-like reference behind the ear provided the best signal-to-noise ratio for AEPs compared to other locations.

16.- More stimulus repetitions were needed to obtain clear dolphin AEPs compared to human studies. Hundreds to a few thousand trials were typical.

17.- Faster stimulus presentation rates of 2-8 stimuli per second reduced the amplitude but preserved the pattern of the dolphin AEPs.

18.- Habituation was observed, with the AEP magnitude decreasing after dolphins listened to the same stimuli over many blocks of trials.

19.- The dolphin AEPs showed a polarity reversal between anterior and posterior electrodes, as previously reported in the early Russian studies.

20.- An average reference across all 16 electrodes provided a cleaner AEP signal than a single mastoid reference or strong low-pass filtering.

21.- Dolphin AEPs recorded with the animal fully submerged were attenuated compared to in-air recordings, likely due to saltwater shunting effects.

22.- A "shallow bite plate" setup with the blowhole and EEG electrodes in air but the lower jaw underwater enabled clearer recordings.

23.- The lecturer aimed to study auditory attention in dolphins using an oddball paradigm with standard, deviant, and target tone stimuli.

24.- Dolphins were trained to respond when they heard a louder target tone and ignore quieter background tones of low and high pitch.

25.- The dolphin AEPs tended to be larger in magnitude for the deviant tones compared to the standard tones in the oddball task.

26.- Preliminary results from one dolphin suggested AEP magnitude was larger when attending to deviants but smaller when ignoring them.

27.- Dolphin AEPs are noisier and more challenging to record than human EEG, but multiple techniques can help improve the signal quality.

28.- In-air recordings are cleaner but unnatural. Fully submerged recordings are natural but prone to saltwater shunting. A compromise is needed.

29.- Other researchers have started to develop wearable EEG loggers for aquatic mammals to enable recording brain activity during natural behaviors.

30.- This dolphin EEG research was a collaborative effort involving government, military, academic and non-profit organizations as well as animal trainers.

Knowledge Vault built byDavid Vivancos 2024