Figure 3 - uploaded by Attila Andics
Content may be subject to copyright.
MVPA using searchlight. A, Brain regions within the visually-responsive cortex of dogs and humans that discriminate conspecific from heterospecific (red) and face from occiput (blue) stimuli. The mean classifier accuracy significance level (p) on each voxel was calculated using permutation testing (see Materials and Methods) p , 0.001 uncorrected and p , 0.05 cluster-corrected for FWE for dogs and p , 0.000001 uncorrected and p , 0.001 cluster corrected for FWE for humans, the searchlight used a spherical kernel with a radius of 4 mm for dogs and 8 mm for humans. B, Histograms depicting classification accuracy across participants for each cluster peak. L = left; R = right; cSSG = caudal ectosylvian gyrus; mSSG = mid suprasylvian gyrus; FuG = fusiform gyrus; IFG = inferior frontal gyrus; IOG = inferior occipital gyrus; ITG = inferior temporal gyrus; MOG = middle occipital gyrus; pMTG = posterior middle temporal gyrus. See also Extended Data Figure 3-1.
Source publication
Conspecific-preference in social perception is evident for multiple sensory modalities and in many species. There is also a dedicated neural network for face processing in primates. Yet, the evolutionary origin and the relative role of neural species-sensitivity and face-sensitivity in visuo-social processing are largely unknown. In this comparativ...
Citations
... To understand how higher order cognition is processed in the dog´s brain, we also need to understand the brain´s fundamental organization and how it processes sensory input at the lower levels. To this end, recent research has mapped out the visual, olfactory and auditory cortices in awake and unrestrained dogs using functional magnetic resonance imaging (fMRI; Andics et al. 2014Andics et al. , 2016Boch et al. 2021Bunford et al. 2020;Cuaya et al. 2016Cuaya et al. , 2022Dilks et al. 2015;Gillette et al. 2022;Jia et al. 2014;Phillips et al. 2022). However, our best understanding of the canine somatosensory cortex dates back to almost 70 years ago (Fritsch & Hitzig, 1870/1963Hamuy et al. 1956), using invasive methods, a small sample and focusing on selected parts of the canine cortex. ...
Dogs are increasingly used as a model for neuroscience due to their ability to undergo functional MRI fully awake and unrestrained, after extensive behavioral training. Still, we know rather little about dogs’ basic functional neuroanatomy, including how basic perceptual and motor functions are localized in their brains. This is a major shortcoming in interpreting activations obtained in dog fMRI. The aim of this preregistered study was to localize areas associated with somatosensory processing. To this end, we touched N = 22 dogs undergoing fMRI scanning on their left and right flanks using a wooden rod. We identified activation in anatomically defined primary and secondary somatosensory areas (SI and SII), lateralized to the contralateral hemisphere depending on the side of touch, and importantly also activation beyond SI and SII, in the cingulate cortex, right cerebellum and vermis, and the sylvian gyri. These activations may partly relate to motor control (cerebellum, cingulate), but also potentially to higher-order cognitive processing of somatosensory stimuli (rostral sylvian gyri), and the affective aspects of the stimulation (cingulate). We also found evidence for individual side biases in a vast majority of dogs in our sample, pointing at functional lateralization of somatosensory processing. These findings not only provide further evidence that fMRI is suited to localize neuro-cognitive processing in dogs, but also expand our understanding of in vivo touch processing in mammals, beyond classically defined primary and secondary somatosensory cortices.
... To understand how higher order cognition is processed in the dog´s brain, we also need to understand the brain´s fundamental organization and how it processes sensory input at the lower levels. To this end, recent research has mapped out the visual, olfactory and auditory cortices in awake and unrestrained dogs using functional magnetic resonance imaging (fMRI; Andics et al., 2014Andics et al., , 2016Boch et al., 2021Boch et al., , 2023Bunford et al., 2020;Cuaya et al., 2016Cuaya et al., , 2022Dilks et al., 2015;Gillette et al., 2022;Jia et al., 2014;Phillips et al., 2022). However, our best understanding of the canine somatosensory cortex dates back to almost 70 years ago (Hamuy et al., 1956), using invasive methods, a small sample and focusing on selected parts of the canine cortex. ...
Dogs are increasingly used as a model for neuroscience due to their ability to undergo functional MRI fully awake and unrestrained, after extensive behavioral training. Still, we know rather little about dogs’ basic functional neuroanatomy, including how basic perceptual and motor functions are localized in their brains. This is a major shortcoming in interpreting activations obtained in dog fMRI. The aim of this preregistered study was to localize areas associated with somatosensory processing. To this end, we touched N = 22 dogs undergoing fMRI scanning on their left and right flanks using a wooden rod. We identified activation in anatomically defined primary and secondary somatosensory areas (SI and SII), lateralized to the contralateral hemisphere depending on the side of touch, as well as activations, beyond an anatomical mask of SI and SII, in the cingulate cortex, right cerebellum and vermis, and the Sylvian gyri. These activations may partly relate to motor control (cerebellum, cingulate), but also potentially to higher-order cognitive processing of somatosensory stimuli (rostral Sylvian gyri), and the affective aspects of the stimulation (cingulate). We also found evidence for individual side biases in a vast majority of dogs in our sample, pointing at functional lateralization of somatosensory processing. These findings not only provide further evidence that fMRI is suited to localize neuro-cognitive processing in dogs in vivo, but also expand our understanding of touch processing in mammals, beyond classically defined primary and secondary somatosensory cortices.
Significance Statement
To understand brain function and evolution, it is necessary to look beyond the human lineage. This study provides insights into the engagement of brain areas related to somatosensation using whole-brain non-invasive neuroimaging of trained, non-sedated, and unrestrained pet dogs. It showcases again the usefulness of non-invasive methods, in particular fMRI, for investigating brain function and advances the mapping of brain functions in dogs; using this non-invasive approach without sedation, we are able to identify previously unknown potential higher-order processing areas and offer a quantification of touch processing lateralization.
... Second, we hypothesized that the dog action observation network, as in humans, includes occipito-temporal brain areas associated with faceand body perception (i.e., agent areas), as well as areas involved in processing of dynamic aspects of social cues and action features [6][7][8] . First evidence suggests that the dog agent areas are housed in the occipito-temporal ectomarginal, the mid and caudal suprasylvian gyrus, but results have been mixed [42][43][44][45][46][47][48][49] . Brain areas associated with the processing of dynamic aspects have yet to be investigated. ...
... Starting with a discussion of the functional analogies, our results show that the AONs of both species include occipital-temporal regions. Observing dogs and humans performing transitive or intransitive actions elicited greater activation and functional connectivity with V1 in the temporal body and agent-sensitive areas of both species (i.e., dog mid and caudal suprasylvian gyrus 46,49 and human inferior temporal cortex 58 ). ...
... In dogs, we found no significant activation increases in response to human compared to conspecific actions. The reversed contrast only led to significantly more activation in the mid suprasylvian animate area, which has been previously associated with higher sensitivity towards conspecifics 46 . The lack of pronounced differences between the perception of dog and human actions may be surprising at first sight. ...
Action observation is a fundamental pillar of social cognition. Neuroimaging research has revealed a human and primate action observation network (AON) encompassing fronto-temporo-parietal areas with links to a species' imitation tendencies and relative lobe expansion. Dogs (Canis familiaris) have good action perception and imitation skills and a less expanded parietal than temporal lobe, but their AON remains unexplored. We conducted a functional MRI study with 28 dogs and 40 humans and found functionally analogous involvement of somatosensory and temporal brain areas of both species' AONs and responses to transitive and intransitive action observation in line with their imitative skills. However, activation and task-based functional connectivity measures suggested significantly less parietal lobe involvement in dogs than in humans. These findings advance our understanding of the neural bases of action understanding and the convergent evolution of social cognition, with analogies and differences resulting from similar social environments and divergent brain expansion, respectively.
... Research so far suggests an involvement of dogs' temporal lobe in face perception, but inconclusive results have triggered a debate on whether the occipito-temporal specialization for face perception in dogs matches that of humans [33][34][35][36][37] . Apart from one electroencephalography (EEG) study 38 , prior neuroimaging studies did not find greater activation for faces compared to scrambled images 33,34 , but compared to scenes 34 or objects 34,36,39 , or didn't have any non-facial controls 35 , questioning if face-sensitivity rather reflects differences in low-level visual properties. ...
... Apart from one electroencephalography (EEG) study 38 , prior neuroimaging studies did not find greater activation for faces compared to scrambled images 33,34 , but compared to scenes 34 or objects 34,36,39 , or didn't have any non-facial controls 35 , questioning if face-sensitivity rather reflects differences in low-level visual properties. Further, almost all prior studies lacked animate stimuli other than faces [33][34][35][36]38,39 and the only study 37 with another animate stimulus category (i.e., the back of the head) had no inanimate control condition. Thus, studies so far could not control for animacy as an alternate explanation of the supposed face-sensitive responses. ...
... Thus, studies so far could not control for animacy as an alternate explanation of the supposed face-sensitive responses. Results regarding species preferences were also mixed, ranging from no conspecific-preference 34 to separate regions for dog and human face perception 35 and a recent report of a conspecific-preferring visual region 37 . Therefore, previous studies carry several limitations that prevent a better understanding of how dogs perceive others compared to humans: they cannot disentangle face-sensitive from general animate vs. inanimate perception, are inconclusive regarding perception of con-and hetero-specific individuals, and provide limited insights into potentially convergent neural underpinnings of face perception, with only two comparative studies so far 37,39 . ...
Comparing the neural correlates of socio-cognitive skills across species provides insights into the evolution of the social brain and has revealed face- and body-sensitive regions in the primate temporal lobe. Although from a different lineage, dogs share convergent visuo-cognitive skills with humans and a temporal lobe which evolved independently in carnivorans. We investigated the neural correlates of face and body perception in dogs (N = 15) and humans (N = 40) using functional MRI. Combining univariate and multivariate analysis approaches, we found functionally analogous occipito-temporal regions involved in the perception of animate entities and bodies in both species and face-sensitive regions in humans. Though unpredicted, we also observed neural representations of faces compared to inanimate objects, and dog compared to human bodies in dog olfactory regions. These findings shed light on the evolutionary foundations of human and dog social cognition and the predominant role of the temporal lobe.
... The dog brain is currently the focus of several projects, including, among others, EEG (Iotchev et al. 2020), brain banking efforts (Urfer et al. 2021;Sandor et al. 2021Sandor et al. , 2022, with fMRI being the most proliferate field in recent years (Aulet et al. 2019;Karl et al. 2021;Prichard et al. 2021). Recent dog fMRI studies are venturing into a wide array of topics, from epilepsy (Beckmann et al. 2021) to comparative face and voice processing (Bunford et al. 2017;Boros et al. 2021;Bunford 2020). ...
Compared to the field of human fMRI, knowledge about functional networks in dogs is scarce. In this paper, we present the first anatomically-defined ROI (region of interest) based functional network map of the companion dog brain. We scanned 33 awake dogs in a “task-free condition”. Our trained subjects, similarly to humans, remain willingly motionless during scanning. Our goal is to provide a reference map with a current best estimate for the organisation of the cerebral cortex as measured by functional connectivity. The findings extend a previous spatial ICA (independent component analysis) study (Szabo et al. in Sci Rep 9(1):1.25. https://doi.org/10.1038/s41598-019-51752-2, 2019), with the current study including (1) more subjects and (2) improved scanning protocol to avoid asymmetric lateral distortions. In dogs, similarly to humans (Sacca et al. in J Neurosci Methods. https://doi.org/10.1016/j.jneumeth.2021.109084, 2021), ageing resulted in increasing framewise displacement (i.e. head motion) in the scanner. Despite the inherently different approaches between model-free ICA and model-based ROI, the resulting functional networks show a remarkable similarity. However, in the present study, we did not detect a designated auditory network. Instead, we identified two highly connected, lateralised multi-region networks extending to non-homotropic regions (Sylvian L, Sylvian R), including the respective auditory regions, together with the associative and sensorimotor cortices and the insular cortex. The attention and control networks were not split into two fully separated, dedicated networks. Overall, in dogs, fronto-parietal networks and hubs were less dominant than in humans, with the cingulate gyrus playing a central role. The current manuscript provides the first attempt to map whole-brain functional networks in dogs via a model-based approach.
... This technology has also shown that different regions of the dog cortex process dog vs. human facial expressions and that these regions (Thompkins et al. 2018(Thompkins et al. , 2021, in dogs seem to be analogous to those found in humans, suggesting the existence of shared ancient neural networks for emotion cue perception (Haxby et al. 2000;Thompkins et al. 2021). Whether dogs have a specific brain region for face processing is less clear, with some fMRI studies finding a dog face region (Cuaya et al. 2016;Dilks et al. 2015;Thompkins et al. 2018), while others do not (Bunford et al. 2020;Szabó et al. 2020). Bunford and colleagues (2020) suggested that the inconsistency of results may be due to sensitivity of analysis, contrasts used and/or data analysis. ...
Comparative studies of human–dog cognition have grown exponentially since the 2000’s, but the focus on how dogs look at us (as well as other dogs) as social partners is a more recent phenomenon despite its importance to human–dog interactions. Here, we briefly summarise the current state of research in visual perception of emotion cues in dogs and why this area is important; we then critically review its most commonly used methods, by discussing conceptual and methodological challenges and associated limitations in depth; finally, we suggest some possible solutions and recommend best practice for future research. Typically, most studies in this field have concentrated on facial emotional cues, with full body information rarely considered. There are many challenges in the way studies are conceptually designed (e.g., use of non-naturalistic stimuli) and the way researchers incorporate biases (e.g., anthropomorphism) into experimental designs, which may lead to problematic conclusions. However, technological and scientific advances offer the opportunity to gather much more valid, objective, and systematic data in this rapidly expanding field of study. Solving conceptual and methodological challenges in the field of emotion perception research in dogs will not only be beneficial in improving research in dog–human interactions, but also within the comparative psychology area, in which dogs are an important model species to study evolutionary processes.
... There is also evidence that they process human faces differently than humans do (e.g. 38,39 ). These differences can have a significant effect on the reciprocity of emotional expressions, a key function of prosody, and suggest that mirroring events (i.e. when mothers and infants display similar facial expression one after another 40,41 ) are probably rare or completely missing from an interaction with a dog. ...
Parents tend to use a specific communication style, including specific facial expressions, when speaking to their preverbal infants which has important implications for children’s healthy development. In the present study, we investigated these facial prosodic features of caregivers with a novel method that compares infant-, dog- and adult-directed communication. We identified three novel facial displays in addition to the already described three facial expressions (i.e. the ‘prosodic faces’) that mothers and fathers are typically displaying when interacting with their 1–18 month-old infants and family dogs, but not when interacting with another adult. The so-called Special Happy expression proved to be the most frequent face type during infant- and dog-directed communication which always includes a Duchenne marker to convey an honest and intense happy emotion of the speaker. These results suggest that the ‘prosodic faces’ play an important role in both adult-infant and human–dog interactions and fulfil specific functions: to call and maintain the partner’s attention, to foster emotionally positive interactions, and to strengthen social bonds. Our study highlights the relevance of future comparative studies on facial prosody and its potential contribution to healthy emotional and cognitive development of infants.
... The use of the same template brain ( Czeibert et al., 2019 ) in this study and in many of our previous studies (e.g. Boros et al., 2020 ;Bálint et al., 2020 ;Bunford et al., 2020 ;Szabó et al., 2020 ) allows for a direct comparison of the coordinates and brain regions. As a result of our several-month-long fMRI training, dogs were able to lie motionless in the scanner (for > 6 min) without restriction. ...
Sine-wave f0 fMRI Dog Human a b s t r a c t Voice-sensitivity in the auditory cortex of a range of mammals has been proposed to be determined primarily by tuning to conspecific auditory stimuli, but recent human findings indicate a role for a more general tuning to voicelikeness. Vocal emotional valence, a central characteristic of vocalisations, has been linked to the same basic acoustic parameters across species. Comparative neuroimaging revealed that during voice perception, such acoustic parameters modulate emotional valence-sensitivity in auditory cortical regions in both family dogs and humans. To explore the role of voicelikeness in auditory emotional valence-sensitivity across species, here we constructed artificial emotional sounds in two sound categories: voice-like vs. sine-wave sounds, parametrically modulating two main acoustic parameters, f0 and call length. We hypothesised that if mammalian auditory systems are characterised by a general tuning to voicelikeness, voice-like sounds will be processed preferentially, and acoustic parameters for voice-like sounds will be processed differently than for sine-wave sounds-both in dogs and humans. We found cortical areas in both species that responded stronger to voice-like than to sine-wave stimuli, while there were no regions responding stronger to sine-wave sounds in either species. Additionally, we found that in bilateral primary and emotional valence-sensitive auditory regions of both species, the processing of voice-like and sine-wave sounds are modulated by f0 in opposite ways. These results reveal functional similarities between evolutionarily distant mammals for processing voicelikeness and its effect on processing basic acoustic cues of vocal emotions.
... As an important signal of sensory input, vision is involved in many processes, such as facial recognition preferences in the brain (Russ and Leopold, 2015), the formation of reward mechanisms (Tsurugizawa et al., 2012), the processing of looming stimulation (Cléry et al., 2020), and 3D shapes. There are some classical brain regions activated (Tsurugizawa et al., 2010), but some other brain regions have been activated and reported in humans (Bunford et al., 2020), suggesting the value of studying the functional organization of the brain with visual stimulation. ...
As a non-radiative, non-invasive imaging technique, functional magnetic resonance imaging (fMRI) has excellent effects on studying the activation of blood oxygen levels and functional connectivity of the brain in human and animal models. Compared with resting-state fMRI, fMRI combined with stimulation could be used to assess the activation of specific brain regions and the connectivity of specific pathways and achieve better signal capture with a clear purpose and more significant results. Various fMRI methods and specific stimulation paradigms have been proposed to investigate brain activation in a specific state, such as electrical, mechanical, visual, olfactory, and direct brain stimulation. In this review, the studies on animal brain activation using fMRI combined with different stimulation methods were retrieved. The instruments, experimental parameters, anesthesia, and animal models in different stimulation conditions were summarized. The findings would provide a reference for studies on estimating specific brain activation using fMRI combined with stimulation.
... Unlike in humans, however, this is less robust because the face selectivity is only evident when a uniform field is used as the baseline but not when scrambled faces are used [86]. Both human and dog observers evince neural activation to conspecifics, but this signal is, again, more selective in humans than in dogs [87]. ...
Studies of face perception in primates elucidate the psychological and neural mechanisms that support this critical and complex ability. Recent progress in characterizing face perception across species, for example in insects and reptiles, has highlighted the ubiquity over phylogeny of this key ability for social interactions and survival. Here, we review the competence in face perception across species and the types of computation that support this behavior. We conclude that the computational complexity of face perception evinced by a species is not related to phylogenetic status and is, instead, largely a product of environmental context and social and adaptive pressures. Integrating findings across evolutionary data permits the derivation of computational principles that shed further light on primate face perception.
