Hubbry Logo
Two-streams hypothesisTwo-streams hypothesisMain
Open search
Two-streams hypothesis
Community hub
Two-streams hypothesis
logo
7 pages, 0 posts
0 subscribers
Be the first to start a discussion here.
Be the first to start a discussion here.
Two-streams hypothesis
Two-streams hypothesis
Two-streams hypothesis
View on Wikipedia
from Wikipedia

The two-streams hypothesis is a model of the neural processing of vision as well as hearing.[1] The hypothesis, given its initial characterisation in a paper by David Milner and Melvyn A. Goodale in 1992, argues that humans possess two distinct visual systems.[2] Recently there seems to be evidence of two distinct auditory systems as well. As visual information exits the occipital lobe, and as sound leaves the phonological network, it follows two main pathways, or "streams". The ventral stream (also known as the "what pathway") leads to the temporal lobe, which is involved with object and visual identification and recognition. The dorsal stream (or, "where pathway") leads to the parietal lobe, which is involved with processing the object's spatial location relative to the viewer and with speech repetition.

History

[edit]
How
What
The dorsal stream (green) and ventral stream (purple) are shown. They originate from a common source in the visual cortex

Several researchers had proposed similar ideas previously. The authors themselves credit the inspiration of work on blindsight by Weiskrantz, and previous neuroscientific vision research. Schneider first proposed the existence of two visual systems for localisation and identification in 1969.[3] Ingle described two independent visual systems in frogs in 1973.[4] Ettlinger reviewed the existing neuropsychological evidence of a distinction in 1990.[5] Moreover, Trevarthen had offered an account of two separate mechanisms of vision in monkeys back in 1968.[6]

In 1982, Ungerleider and Mishkin distinguished the dorsal and ventral streams, as processing spatial and visual features respectively, from their lesion studies of monkeys – proposing the original where vs what distinction.[7] Though this framework was superseded by that of Milner & Goodale, it remains influential.[8]

One hugely influential source of information that has informed the model has been experimental work exploring the extant abilities of visual agnosic patient D.F. The first, and most influential report, came from Goodale and colleagues in 1991[9] and work is still being published on her two decades later.[10] This has been the focus of some criticism of the model due to the perceived over-reliance on findings from a single case.

Two visual systems

[edit]

Goodale and Milner[2] amassed an array of anatomical, neuropsychological, electrophysiological, and behavioural evidence for their model. According to their data, the ventral 'perceptual' stream computes a detailed map of the world from visual input, which can then be used for cognitive operations, and the dorsal 'action' stream transforms incoming visual information to the requisite egocentric (head-centered) coordinate system for skilled motor planning. The model also posits that visual perception encodes spatial properties of objects, such as size and location, relative to other objects in the visual field; in other words, it utilizes relative metrics and scene-based frames of reference. Visual action planning and coordination, on the other hand, uses absolute metrics determined via egocentric frames of reference, computing the actual properties of objects relative to the observer. Thus, grasping movements directed towards objects embedded in size-contrast-ambiguous scenes have been shown to escape the effects of these illusions, as different frames of references and metrics are involved in the perception of the illusion versus the execution of the grasping act.[11]

Norman[12] proposed a similar dual-process model of vision, and described eight main differences between the two systems consistent with other two-system models.

Factor Ventral system (what) Dorsal system (how)
Function Recognition/identification Visually guided behaviour
Sensitivity High spatial frequencies - details High temporal frequencies - motion
Memory Long-term stored representations Only very short-term storage
Speed Relatively slow Relatively fast
Consciousness Typically high Typically low
Frame of reference Allocentric or object-centered Egocentric or viewer-centered
Visual input Mainly foveal or parafoveal Across retina
Monocular vision Generally reasonably small effects Often large effects e.g. motion parallax

Dorsal stream

[edit]
Connections of dorsal visual stream

The dorsal stream is proposed to be involved in the guidance of actions and recognizing where objects are in space. The dorsal stream projects from the primary visual cortex to the posterior parietal cortex. It was initially termed the "where" pathway since it was thought that the dorsal stream processes information regarding the spatial properties of an object.[13] However, later research conducted on a famous neuropsychological patient, Patient D.F., revealed that the dorsal stream is responsible for processing the visual information needed to construct the representations of objects one wishes to manipulate. Those findings led the nickname of the dorsal stream to be updated to the "how" pathway.[14][15] The dorsal stream is interconnected with the parallel ventral stream (the "what" stream) which runs downward from V1 into the temporal lobe.

General features

[edit]

The dorsal stream is involved in spatial awareness and guidance of actions (e.g., reaching). In this it has two distinct functional characteristics—it contains a detailed map of the visual field, and is also good at detecting and analyzing movements.

The dorsal stream commences with purely visual functions in the occipital lobe before gradually transferring to spatial awareness at its termination in the parietal lobe.

The posterior parietal cortex is essential for "the perception and interpretation of spatial relationships, accurate body image, and the learning of tasks involving coordination of the body in space".[16]

It contains individually functioning lobules. The lateral intraparietal sulcus (LIP) contains neurons that produce enhanced activation when attention is moved onto the stimulus or the animal saccades towards a visual stimulus, and the ventral intraparietal sulcus (VIP) where visual and somatosensory information are integrated.

Effects of damage or lesions

[edit]

Damage to the posterior parietal cortex causes a number of spatial disorders including:

  • Simultanagnosia: where the patient can only describe single objects without the ability to perceive it as a component of a set of details or objects in a context (as in a scenario, e.g. the forest for the trees).
  • Optic ataxia: where the patient cannot use visuospatial information to guide arm movements.
  • Hemispatial neglect: where the patient is unaware of the contralesional half of space (that is, they are unaware of things in their left field of view and focus only on objects in the right field of view; or appear unaware of things in one field of view when they perceive them in the other). For example, a person with this disorder may draw a clock, and then label all twelve of the numbers on one side of the face and consider the drawing complete.
  • Akinetopsia: inability to perceive motion.
  • Apraxia: inability to produce discretionary or volitional movement in the absence of muscular disorders.

Ventral stream

[edit]
Ventromedial view of visual stream
Ventolateral view of visual stream

The ventral stream is associated with object recognition and form representation. Also described as the "what" stream, it has strong connections to the medial temporal lobe (which is associated with long-term memories), the limbic system (which controls emotions), and the dorsal stream (which deals with object locations and motion).

The ventral stream gets its main input from the parvocellular (as opposed to magnocellular) layer of the lateral geniculate nucleus of the thalamus. These neurons project to V1 sublayers 4Cβ, 4A, 3B and 2/3a[17] successively. From there, the ventral pathway goes through V2 and V4 to areas of the inferior temporal lobe: PIT (posterior inferotemporal), CIT (central inferotemporal), and AIT (anterior inferotemporal). Each visual area contains a full representation of visual space. That is, it contains neurons whose receptive fields together represent the entire visual field. Visual information enters the ventral stream through the primary visual cortex and travels through the rest of the areas in sequence.

Moving along the stream from V1 to AIT, receptive fields increase their size, latency, and the complexity of their tuning. For example, recent studies have shown that the V4 area is responsible for color perception in humans, and the V8 (VO1) area is responsible for shape perception, while the VO2 area, which is located between these regions and the parahippocampal cortex, integrates information about the color and shape of stimuli into a holistic image.[18]

All the areas in the ventral stream are influenced by extraretinal factors in addition to the nature of the stimulus in their receptive field. These factors include attention, working memory, and stimulus salience. Thus the ventral stream does not merely provide a description of the elements in the visual world—it also plays a crucial role in judging the significance of these elements.

Damage to the ventral stream can cause inability to recognize faces or interpret facial expression.[19]

Two auditory systems

[edit]

Auditory ventral stream

[edit]
comprehension of the phrase 'my cat' in the extended version of Hickok and Poeppel's dual pathway model

Along with the visual ventral pathway being important for visual processing, there is also a ventral auditory pathway emerging from the primary auditory cortex.[20] In this pathway, phonemes are processed posteriorly to syllables and environmental sounds.[21] The information then joins the visual ventral stream at the middle temporal gyrus and temporal pole. Here the auditory objects are converted into audio-visual concepts.[22]

Auditory dorsal stream

[edit]

The function of the auditory dorsal pathway is to map the auditory sensory representations onto articulatory motor representations. Hickok & Poeppel claim that the auditory dorsal pathway is necessary because, "learning to speak is essentially a motor learning task. The primary input to this is sensory, speech in particular. So, there must be a neural mechanism that both codes and maintains instances of speech sounds, and can use these sensory traces to guide the tuning of speech gestures so that the sounds are accurately reproduced."[23]

repetition of the phrase 'what is your name?' in the extended version of Hickok and Poeppel's dual pathway model

In contrast to the ventral stream's auditory processing, information enters from the primary auditory cortex into the posterior superior temporal gyrus and posterior superior temporal sulcus. From there the information moves to the beginning of the dorsal pathway, which is located at the boundary of the temporal and parietal lobes near the Sylvian fissure. The first step of the dorsal pathway begins in the sensorimotor interface, located in the left Sylvian parietal temporal (Spt) (within the Sylvian fissure at the parietal-temporal boundary). The spt is important for perceiving and reproducing sounds. This is evident because its ability to acquire new vocabulary, be disrupted by lesions and auditory feedback on speech production, articulatory decline in late-onset deafness and the non-phonological residue of Wernicke's aphasia; deficient self-monitoring. It is also important for the basic neuronal mechanisms for phonological short-term memory. Without the Spt, language acquisition is impaired. The information then moves onto the articulatory network, which is divided into two separate parts. The articulatory network 1, which processes motor syllable programs, is located in the left posterior inferior temporal gyrus and Brodmann's area 44 (pIFG-BA44).[24] The articulatory network 2 is for motor phoneme programs and is located in the left M1-vBA6.[25]

Conduction aphasia affects a subject's ability to reproduce speech (typically by repetition), though it has no influence on the subject's ability to comprehend spoken language. This shows that conduction aphasia must reflect not an impairment of the auditory ventral pathway but instead of the auditory dorsal pathway. Buchsbaum et al[26] found that conduction aphasia can be the result of damage, particularly lesions, to the Spt (Sylvian parietal temporal). This is shown by the Spt's involvement in acquiring new vocabulary, for while experiments have shown that most conduction aphasiacs can repeat high-frequency, simple words, their ability to repeat low-frequency, complex words is impaired. The Spt is responsible for connecting the motor and auditory systems by making auditory code accessible to the motor cortex. It appears that the motor cortex recreates high-frequency, simple words (like cup) in order to more quickly and efficiently access them, while low-frequency, complex words (like Sylvian parietal temporal) require more active, online regulation by the Spt. This explains why conduction aphasiacs have particular difficulty with low-frequency words which requires a more hands-on process for speech production. "Functionally, conduction aphasia has been characterized as a deficit in the ability to encode phonological information for production," namely because of a disruption in the motor-auditory interface.[27] Conduction aphasia has been more specifically related to damage of the arcuate fasciculus, which is vital for both speech and language comprehension, as the arcuate fasiculus makes up the connection between Broca and Wernicke's areas.[28]

Criticisms

[edit]

Goodale & Milner's innovation was to shift the perspective from an emphasis on input distinctions, such as object location versus properties, to an emphasis on the functional relevance of vision to behaviour, for perception or for action. Contemporary perspectives however, informed by empirical work over the past two decades, offer a more complex account than a simple separation of function into two-streams.[29] Recent experimental work for instance has challenged these findings, and has suggested that the apparent dissociation between the effects of illusions on perception and action is due to differences in attention, task demands, and other confounds.[30][31] There are other empirical findings, however, that cannot be so easily dismissed which provide strong support for the idea that skilled actions such as grasping are not affected by pictorial illusions.[32][33][34][35]

Moreover, recent neuropsychological research has questioned the validity of the dissociation of the two streams that has provided the cornerstone of evidence for the model. The dissociation between visual agnosia and optic ataxia has been challenged by several researchers as not as strong as originally portrayed; Hesse and colleagues demonstrated dorsal stream impairments in patient DF;[36] Himmelbach and colleagues reassessed DF's abilities and applied more rigorous statistical analysis demonstrating that the dissociation was not as strong as first thought.[10]

A 2009 review of the accumulated evidence for the model concluded that whilst the spirit of the model has been vindicated the independence of the two streams has been overemphasised.[37] Goodale & Milner themselves have proposed the analogy of tele-assistance, one of the most efficient schemes devised for the remote control of robots working in hostile environments. In this account, the dorsal stream is viewed as a semi-autonomous function that operates under guidance of executive functions which themselves are informed by ventral stream processing.[38]

Thus the emerging perspective within neuropsychology and neurophysiology is that, whilst a two-systems framework was a necessary advance to stimulate study of the highly complex and differentiated functions of the two neural pathways; the reality is more likely to involve considerable interaction between vision-for-action and vision-for-perception. Robert McIntosh and Thomas Schenk summarize this position as follows:

We should view the model not as a formal hypothesis, but as a set of heuristics to guide experiment and theory. The differing informational requirements of visual recognition and action guidance still offer a compelling explanation for the broad relative specializations of dorsal and ventral streams. However, to progress the field, we may need to abandon the idea that these streams work largely independently of one other, and to address the dynamic details of how the many visual brain areas arrange themselves from task to task into novel functional networks.[39]

— Thomas Schenk and Robert D. McIntosh, "Do We Have Independent Visual Streams for Perception and Action?"

See also

[edit]

References

[edit]
Revisions and contributorsEdit on WikipediaRead on Wikipedia

Two-streams hypothesis

View on Grokipedia
from Grokipedia
The two-streams hypothesis, a foundational model in visual neuroscience, proposes that the processes visual information through two parallel cortical pathways originating from primary : the ventral stream, which extends into the and supports object identification and perceptual representation (often termed the "what" pathway), and the dorsal stream, which projects to the and facilitates spatial awareness and visually guided actions (known as the "where" or "how" pathway). This highlights functional specialization in the , where perception and action are dissociated to enable efficient interaction with the environment. The hypothesis originated from lesion studies in monkeys conducted by Leslie G. Ungerleider and Mortimer Mishkin in the early 1980s, who observed that damage to the impaired but spared spatial abilities, while parietal lesions disrupted localization without affecting identification. Their work built on earlier neuroanatomical evidence of segregated projections from striate cortex (V1) to inferotemporal and posterior parietal regions, establishing the streams as distinct multisynaptic pathways. In 1992, Melvyn A. Goodale and A. David Milner refined the model based on human patient data, particularly from D.F., who suffered ventral stream damage yet could perform accurate actions on objects despite profound perceptual deficits, shifting emphasis from "where" to "how" for the dorsal pathway. Subsequent and electrophysiological studies in both monkeys and humans have corroborated the , revealing hierarchical within each stream: the ventral pathway involves areas like V4 and inferotemporal cortex for feature integration and category-specific recognition, while the dorsal stream includes regions such as the lateral intraparietal area () and medial superior temporal area (MST) for motion and depth critical to . Evidence from functional MRI shows stream-specific activation patterns during tasks requiring versus grasping, underscoring their independence yet interconnected nature through feedback loops. The two-streams model has profoundly influenced , informing theories of visuomotor coordination, syndromes, and even computational vision algorithms that mimic biological separation of recognition and . Challenges include debates over stream interactions and whether the dorsal pathway includes substreams for different action types, but it remains a cornerstone for understanding organization.

Overview

Core Concepts

The proposes that the processes sensory , particularly visual input, through parallel cortical pathways that specialize in distinct functions. In its core formulation for vision, one pathway—the ventral stream—handles the identification and recognition of objects ("what" ), while the other—the dorsal stream—manages spatial awareness and guidance of actions ("where" or "how" ). This division allows for efficient, segregated handling of perceptual and motor demands, emerging from the initial convergence of visual signals in primary before diverging into these streams. The hypothesis was originally articulated by Ungerleider and Mishkin in 1982, based on anatomical and behavioral evidence from rhesus monkeys, which demonstrated dissociable pathways projecting from occipital cortex to temporal and parietal regions, respectively. This model emphasized functional specialization, with the temporal pathway supporting object discrimination and the parietal pathway aiding visuospatial tasks. Subsequent refinement by Goodale and Milner in 1992 shifted the emphasis from purely spatial versus object processing to a sharper distinction between conscious (ventral stream) and unconscious, real-time visuomotor control (dorsal stream), drawing on neuropsychological cases of dissociated visual deficits. This functional dissociation underscores the hypothesis's explanatory power: the ventral stream enables deliberate, recognition-based that integrates with and , whereas the dorsal stream operates rapidly and automatically to support immediate actions like grasping or , often bypassing conscious . Analogous parallel streams have been hypothesized for auditory processing, where ventral pathways focus on sound identification and dorsal ones on spatial localization and action integration, extending the model's principles beyond vision.

Anatomical Foundations

The two-streams hypothesis posits a segregation of visual processing into parallel cortical pathways originating from the primary (V1), with distinct anatomical trajectories supported by neuroanatomical tracing studies in . This segregation begins early in the visual system, influenced by inputs from the (LGN) of the , where magnocellular layers predominantly project to regions that feed the dorsal stream, emphasizing motion and spatial sensitivity, while parvocellular layers primarily innervate areas contributing to the ventral stream, focusing on color and form details. Although both streams receive mixed inputs, this differential emphasis from LGN layers 1-2 (magnocellular) and 3-6 (parvocellular) establishes the foundational parallel processing architecture. The ventral stream pathway follows an occipitotemporal route, progressing from V1 through the ventral portions of V2 and V4 to the inferotemporal cortex (IT), particularly areas TE and TEO. These connections form a hierarchical series of corticocortical projections, with V4 serving as a key intermediate station that receives color- and shape-selective inputs before relaying to IT, where neurons exhibit invariance to object position and size. Anatomical studies using anterograde tracers in monkeys confirm dense, reciprocal linkages along this path, underscoring its role as a segregated conduit for detailed visual analysis. In contrast, the dorsal stream pathway takes an occipitoparietal course, extending from V1 via the dorsal aspects of V2 and V3 to the motion-sensitive middle temporal area (MT/V5), and thence to the posterior parietal cortex (PPC), including regions such as the lateral intraparietal area () and ventral intraparietal area (VIP). This route is characterized by broader, more divergent projections, with MT receiving strong inputs from V1's thick stripes in V2 and projecting forward to parietal areas via area V3A. Tracer studies reveal that these connections maintain spatial and motion-biased selectivity, forming a parallel hierarchy distinct from the ventral pathway. While the streams operate in parallel from their early divergence at V1, evidence from connectivity mapping indicates limited cross-stream interactions at higher cortical levels, such as bidirectional projections between IT area TEO and PPC's , which may allow for integration without fully merging the pathways. This anatomical arrangement preserves the hypothesis's core principle of functional specialization through segregated processing routes.

Historical Development

Origins in Visual Neuroscience

The foundations of the two-streams hypothesis emerged from mid-20th-century studies elucidating the functional organization of the . In the 1960s, David Hubel and conducted seminal electrophysiological recordings in the primary (V1) of cats and monkeys, identifying specialized "feature detectors" such as simple cells responsive to oriented and complex cells integrating motion and directionality. Their work demonstrated a hierarchical progression from basic in V1 to more complex feature integration in higher areas, suggesting segregated pathways for processing different aspects of visual information. This laid critical groundwork for understanding how visual inputs diverge into specialized streams beyond V1. Preceding the formal hypothesis, lesion studies in the late 1960s provided early evidence for dual visual processing mechanisms. George Schneider's 1969 experiments on hamsters revealed that tectal lesions (targeting the ) selectively impaired visuomotor orienting and localization, while lesions to the disrupted pattern discrimination and without affecting spatial guidance. These dissociations indicated two anatomically and functionally distinct systems: one for reflexive localization via subcortical structures and another for perceptual analysis in cortical areas, influencing subsequent models of visual pathway segregation in mammals. The two-streams hypothesis was explicitly formulated in 1982 by Leslie G. Ungerleider and Mortimer Mishkin, building on these influences through targeted lesion experiments in rhesus monkeys. They proposed a ventral stream projecting from occipital cortex to the inferior for "what" processing—enabling object identification and form —and a dorsal stream to the posterior parietal cortex for "where" processing—supporting spatial localization and visually guided actions. Monkeys with temporal lesions exhibited deficits in recognizing objects across views or delays, whereas parietal lesions caused impairments in reaching toward or discriminating locations of stimuli, confirming the streams' dissociable roles. This primate-based framework began transitioning to human applications in the late 1980s with the emergence of (PET) imaging, which enabled noninvasive mapping of activation. Early PET studies, such as those demonstrating retinotopic organization in human V1 during visual stimulation, provided initial correlates for segregated processing pathways analogous to the monkey model. These findings marked the onset of functional neuroimaging's role in validating the hypothesis across species, highlighting occipital-parietal and occipital-temporal activations during spatial versus object tasks.

Key Experimental Milestones

In the early 1990s, a significant refinement to the two-streams hypothesis emerged from studies of patient D.F., who suffered from visual form due to damage in the ventral stream following . D.F. exhibited intact visuomotor abilities, such as accurately grasping objects despite being unable to consciously perceive their shape or orientation, demonstrating a dissociation between perception and action. This evidence prompted A. David Milner and Melvyn A. Goodale to propose in their 1992 paper that the dorsal stream is specialized for "how" to interact with objects in real-time visuomotor control, rather than merely "where" they are located, building on earlier lesion studies while emphasizing functional independence between the streams; they elaborated on this in their 1995 book The Visual Brain in Action. Concurrent electrophysiological recordings in awake monkeys during the provided cellular-level validation of stream-specific processing. In the ventral stream, single-unit recordings from the inferotemporal cortex (IT) revealed neurons highly selective for complex object features, such as shapes and textures, supporting its role in independent of spatial location or action demands. Complementing this, recordings in the dorsal stream's middle temporal area (MT) confirmed sensitivity to motion direction and speed, while posterior parietal cortex (PPC) neurons, particularly in areas like the anterior intraparietal area (AIP), responded to visual cues for grasping and reaching, such as object size and orientation for preshaping the hand. Early functional magnetic resonance imaging (fMRI) studies in humans during the late 1990s and early 2000s further corroborated stream segregation by examining brain activation during perception versus action tasks. For instance, when participants viewed objects for identification (engaging the ventral stream), activation was prominent in lateral occipital complex regions, whereas grasping tasks elicited stronger responses in dorsal areas like the superior parietal lobule and intraparietal sulcus, with minimal ventral involvement. These patterns highlighted how the streams process visual information differently based on task demands, aligning with primate data and extending the hypothesis to human cognition.

Visual Streams

Ventral Stream: Perception and Object Recognition

The ventral stream, often referred to as the "what" pathway, plays a central role in visual perception by enabling the identification and categorization of objects through a hierarchical progression of neural processing. This pathway begins in the primary visual cortex (V1) and extends through secondary areas like V2, integrating basic features such as edges and orientations, before advancing to area V4, where more complex attributes like color, form, and texture are combined to form intermediate representations of object surfaces and shapes. In V4, neurons exhibit selectivity for contour fragments and surface properties, facilitating the parsing of object boundaries from backgrounds and contributing to the synthesis of coherent visual forms essential for recognition. This integration culminates in the inferior temporal (IT) cortex, where neurons achieve viewpoint- and size-invariant representations of entire objects, allowing for robust identification regardless of retinal position or transformations. Behavioral manifestations of ventral stream processing are evident in specialized tasks requiring . For instance, face recognition relies heavily on the (FFA) within the ventral occipitotemporal cortex, where neurons respond selectively to facial configurations, supporting the rapid discrimination of individual identities. Similarly, word reading engages the (VWFA) in the left , a region tuned to orthographic strings and invariant to font or case variations, enabling efficient decoding of written language as a perceptual category. Scene categorization, in turn, activates the parahippocampal place area (PPA), which processes holistic spatial layouts and environmental contexts to distinguish indoor from outdoor settings or natural from urban scenes. The ventral stream functions as a slower, more deliberative pathway compared to its dorsal counterpart, with neural latencies typically ranging from 200 to 300 ms post-stimulus onset in human magnetoencephalography (MEG) recordings, reflecting the time required for hierarchical feature integration and recurrent processing that supports conscious awareness, semantic interpretation, and integration with long-term memory. This temporal profile allows the stream to contribute to enduring perceptual representations, such as linking object identities to stored knowledge, rather than immediate visuomotor responses.

Dorsal Stream: Action and Spatial Processing

The dorsal stream, extending from the to the posterior parietal cortex (PPC), specializes in visuomotor transformations that enable egocentric spatial processing and the guidance of actions, such as localizing objects relative to the observer and coordinating movements in real time. This pathway supports the "how" aspect of vision by integrating dynamic visual cues to direct behaviors without relying on conscious object identification. Key inputs to the dorsal stream include motion signals, which facilitate rapid adjustments to changing environments. The dorsal stream comprises two main subdivisions: the dorsomedial stream and the dorsolateral stream, each contributing to distinct aspects of spatial and action processing. The dorsomedial stream, involving areas like V6A in the and the medial intraparietal area (MIP), focuses on the online control of reaching and grasping movements, integrating visual form and motion for arm trajectories. In contrast, the dorsolateral stream encompasses the anterior intraparietal area (AIP) for precise hand shaping during manipulation and the lateral intraparietal area (LIP) for broader spatial navigation and attention shifts. These subdivisions ensure segregated yet interconnected processing for complex visuomotor tasks. Central to dorsal stream function is real-time motion analysis in area MT/V5, which detects object trajectories and feeds into parietal regions for immediate action planning. The PPC then executes coordinate transformations, remapping retinotopic visual inputs into somatotopic frames for hand-eye coordination, allowing seamless alignment of gaze and limb movements. This process underpins behaviors like visually guided saccades, where neurons encode target locations to direct rapid eye shifts toward salient stimuli. Behavioral evidence highlights the dorsal 's role in practical actions, such as obstacle avoidance during reaching, where automatic trajectory corrections prevent collisions based on spatial layout rather than object details. Patients with dorsal lesions, like those exhibiting , fail to adapt hand paths around barriers, underscoring the pathway's necessity for implicit visuomotor guidance. Similarly, tool use demonstrates AIP's involvement in grip preshaping for functional manipulation—such as wielding a —without explicit awareness of the tool's identity, as observed in individuals with ventral damage who perform actions accurately despite perceptual deficits.

Extensions to Auditory Processing

Auditory Ventral Stream

The auditory ventral stream represents the auditory counterpart to the visual ventral stream, emphasizing the identification and semantic interpretation of sounds rather than their spatial attributes. This pathway processes auditory "what" information, enabling the recognition of sound objects and their associated meanings, much like object recognition in the visual domain. Proposed as an extension of the two-streams hypothesis to audition, it highlights modality-specific adaptations for perceiving complex acoustic environments. The pathway originates in the primary auditory cortex (A1) and proceeds anteriorly through belt regions such as the middle lateral (ML) and anterolateral (AL) areas, then along the superior temporal gyrus (STG) toward the anterior temporal lobe and ventrolateral prefrontal cortex (vlPFC). This hierarchical progression allows for increasingly abstract representations, with early stages encoding basic acoustic features and later regions integrating contextual and semantic information. In humans, the stream projects ventrolaterally from anterior and middle STG to middle and inferior temporal cortices, facilitating interfaces between sound processing and higher cognitive networks. Core functions of the auditory ventral stream include , environmental sound categorization, and phonological processing. In speech, the STG exhibits a : middle regions handle phonemes, anterior-superior areas process words, and the most anterior parts manage phrases, supporting comprehension through invariant representations of temporally complex sounds. For environmental sounds, AL regions categorize stimuli like vocalizations despite acoustic variability, contributing to scene analysis. Phonological processing occurs early, with AL neurons showing categorical responses to speech contrasts, such as distinguishing in ambiguous tokens. Overall, the stream maps acoustic input to lexical-semantic meanings, essential for and auditory object identification. Evidence from a supports the ventral stream's role in "what" tasks, showing segregated activation patterns that align with processing specialization, with nonspatial (identity) tasks primarily activating regions and spatial (location) tasks activating parietal regions, indicating functional independence of the auditory ventral pathway in humans.

Auditory Dorsal Stream

The auditory dorsal stream, analogous to its visual counterpart, processes spatial aspects of sound, originating in the core including primary auditory cortex (A1) within Heschl's gyrus and extending through the lateral belt and parabelt regions of the (STG). From these posterior superior temporal areas, the pathway projects dorsally to the (IPL), particularly the and extensions, and further to frontal regions such as the and posterior . This fronto-parietal trajectory facilitates the integration of auditory spatial information with motor systems, distinguishing it from the ventral stream's focus on sound identification. Key functions of the auditory dorsal stream include sound localization, which relies on extracting binaural cues to determine a sound source's position in three-dimensional space. Neurons along this pathway, particularly in the posterior STG and IPL, are tuned to interaural time differences (ITDs) and interaural level differences (ILDs), enabling precise encoding of azimuth (horizontal position) relative to the listener's head. For elevation (vertical position), spectral cues from pinna filtering are processed in parietal regions, with human neuroimaging showing activation in the IPL during tasks involving vertical sound motion. Beyond localization, the dorsal stream supports auditory attention shifts and sensorimotor integration, guiding orienting responses such as head turning or eye movements toward sound sources. Functional MRI studies demonstrate IPL and premotor activation during tasks requiring spatial attention to sounds, linking auditory input to motor planning for actions like vocalization or in response to auditory events. This sensorimotor role extends to coordinating auditory-motor mappings, as evidenced by electrophysiology where dorsal stream neurons respond to both sound location and associated movements.

Supporting Evidence

Neuroimaging and Functional Studies

(fMRI) studies have provided robust evidence for the segregation of the ventral and dorsal visual streams in healthy brains by demonstrating selective activations during tasks emphasizing versus action. For instance, when participants viewed objects for recognition purposes, ventral stream regions such as the lateral occipital complex (LOC) showed heightened activation, whereas dorsal stream areas like the anterior intraparietal sulcus (AIP) exhibited minimal response. In contrast, during visually guided grasping tasks with real 3D objects, AIP and other dorsal regions were preferentially activated, while LOC activity was suppressed, highlighting a functional dissociation between streams for object and manipulation. These findings from early research, including seminal work by Culham and colleagues, underscore how task demands can isolate stream-specific processing in non-invasive group studies. Electroencephalography (EEG) and magnetoencephalography (MEG) have further elucidated the temporal dynamics of the two streams through event-related potentials (ERPs) and fields, revealing differences in processing speed for motion and form features. Dorsal stream responses to dynamic visual stimuli, such as motion onset, emerge rapidly around 100 ms post-stimulus, as indexed by early positive components over posterior parietal electrodes, reflecting quick spatial and action-oriented computations. Ventral stream processing of static form and object identity, however, manifests later, typically 150-200 ms after stimulus presentation, with components like the N170 over occipitotemporal sites supporting perceptual categorization. These temporal distinctions, observed in healthy subjects during simple visual discrimination tasks, support the of parallel but temporally offset pathways, with dorsal responses preceding ventral ones to facilitate rapid integration for . Recent advancements in the , particularly resting-state fMRI, have revealed intrinsic connectivity patterns that delineate stream-specific networks while also indicating cross-modal influences. Analyses of spontaneous BOLD fluctuations show stronger within-network coherence in ventral regions (e.g., ) linked to object semantics and in dorsal areas (e.g., superior parietal lobule) associated with spatial attention, confirming segregated functional architectures even without external stimuli. Moreover, these studies highlight inter-stream and cross-modal connections, such as between visual dorsal areas and auditory motion-sensitive regions, suggesting dynamic interactions that extend the two-streams model beyond isolated visual processing. Such connectivity profiles, derived from large-scale datasets, provide evidence for the hypothesis's applicability in understanding integrated sensory-motor functions in typical brains.

Lesion and Behavioral Evidence

Lesions to the ventral stream, which is associated with and perception, often result in , where patients exhibit profound deficits in identifying and recognizing visual forms despite intact basic visual abilities such as acuity and . A seminal case is that of patient D.F., who suffered bilateral damage to the lateral occipital complex following in 1988; she demonstrated visual form agnosia, performing at chance levels (around 50% accuracy) on tasks requiring explicit judgments of object orientation, width, or shape, such as matching the slant of a slot or the width of Efron blocks. However, D.F. showed preserved visuomotor abilities, accurately preshaping her hand grip to match the width or orientation of objects during grasping tasks, with maximum grip aperture scaling appropriately to object size (e.g., errors less than 5 mm deviation from controls). This dissociation provided key evidence for the functional independence of the ventral stream in conscious perception from the dorsal stream's role in action guidance. In contrast, lesions to the dorsal stream, typically from or trauma affecting the and , lead to , characterized by impaired visually guided reaching and pointing while object perception remains relatively intact. Patients with unilateral parietal lesions, such as those studied in cases of posterior parietal , can accurately describe object features like , , and verbally or through perceptual matching tasks (e.g., achieving over 90% accuracy in object identification), but exhibit significant errors in manual actions, with reaching deviations up to 10-15 cm from target locations under visual guidance. For instance, in a series of patients with right parietal damage, visuomotor coordination was disrupted in the contralesional field, leading to misreaching and grasping inaccuracies, yet discrimination of the same stimuli in non-action tasks was unimpaired. These findings underscore the dorsal stream's specialized role in transforming visual information into spatial-motor coordinates for action. Behavioral paradigms further illustrate dorsal stream deficits through targeted experiments revealing dissociations in spatial and temporal processing. In Efron's delayed matching task, adapted for visual stimuli, patients with parietal lesions show impairments in judging the simultaneity of events occurring at different spatial locations, with thresholds for detecting asynchrony elevated by 50-100 ms compared to controls, while judgments of temporal order for colocalized stimuli remain intact. This deficit, observed in individuals with right posterior parietal damage, highlights the dorsal stream's involvement in integrating spatial with , as the task requires binding location-specific visual inputs over time without relying on . Such paradigms complement lesion studies by isolating dorsal functions in healthy and impaired populations alike.

Criticisms and Modern Perspectives

Challenges to Stream Independence

While the two-streams hypothesis posits a largely independent segregation of the ventral and dorsal visual pathways, neuroanatomical and functional studies have revealed substantial bidirectional interactions that challenge this modularity. Evidence from diffusion tensor imaging indicates the existence of pathways, such as the vertical occipital fasciculus, that facilitate cross-talk between the streams, supporting the integration of (ventral) with spatial action guidance (dorsal). For instance, ventral stream feedback to dorsal areas enables object-specific grasping adjustments, as demonstrated in models accounting for neural dynamics during perception-action tasks where bidirectional connections best explain observed behaviors. further supports this, showing ventral contributions to dorsal processing in tasks requiring object-guided actions, such as reaching, where perceptual details from the ventral stream modulate motor planning in the dorsal stream. Recent psychophysical reexaminations have intensified critiques of stream independence, particularly through scrutiny of classic experiments underpinning the . A 2024 analysis revisited Milner and Goodale's foundational studies, arguing that visual illusions affect action to a greater degree than previously reported, with confounds like and task timing potentially inflating perceived dissociations between and action. These critiques highlight shared representational processes across streams in attention-demanding tasks, where a single underlying mechanism—rather than fully distinct systems—can account for both perceptual reports and motor responses, as evidenced by non-replications of key illusion effects and Bayesian modeling favoring integrated processing. Ongoing debates between proponents like Milner and Goodale and critics emphasize that while distinct representations may exist (two-reps view), the evidence for entirely separate processing systems (two-systems) remains moderate at best. The overlap of impairments in neurodegenerative disorders like Alzheimer's disease further undermines the notion of strict stream independence, revealing concurrent dysfunctions that suggest intertwined dependencies. In Alzheimer's, both ventral stream deficits—such as impaired object and face recognition—and dorsal stream impairments—like reduced motion perception and spatial navigation—are evident early, linked to amyloid deposition and atrophy in shared visual cortices (e.g., Brodmann area 19). Retinal thinning and optic nerve changes correlate with declines in both pathways, as shown in tasks engaging visuospatial organization that recruit elements from each stream, indicating that pathological processes do not respect modular boundaries. This widespread involvement challenges the hypothesis's modularity by demonstrating how unified neurodegenerative mechanisms disrupt the purportedly segregated functions of perception and action.

Recent Developments and Refinements

A 2022 computational study used recurrent neural networks to examine the two-streams hypothesis, finding that ventral stream-like models require longer memory for viewpoint-invariant object classification compared to dorsal stream models for orientation and size estimation tasks. A 2025 study using intracranial EEG during memory-guided actions found increased alpha power (8-13 Hz) in both dorsal () and ventral (ventral temporal cortex) streams during delays, with theta oscillations (2-7 Hz) in the ventral stream. Cross-frequency interactions between these regions indicated coupled stream activity, suggesting the streams operate in an integrated rather than strictly parallel manner during . Proposals for refinements include subdividing the dorsal stream into multiple sub-streams and incorporating cross-sensory integrations to address limitations in unimodal processing. Neuroanatomical studies propose a dorso-dorsal sub-stream for online visuomotor control and a ventro-dorsal sub-stream for semantic action understanding, supported by connectivity patterns in primate parietal cortex that differentiate reaching (dorodorsal) from object manipulation (ventrodorsal) tasks. Additionally, visuo-auditory integrations in the posterior parietal cortex (PPC) enable multisensory binding, as evidenced by synchronized neural responses to congruent stimuli, enhancing spatial localization accuracy by 40% in behavioral paradigms and extending the two-streams framework to multimodal .

References

  1. https://www.cns.nyu.edu/~tony/vns/readings/ungerleider-mishkin-1982.pdf
  2. https://cnbc.cmu.edu/braingroup/papers/goodale_milner_1992.pdf
  3. https://www.bps.org.uk/psychologist/one-brain-two-visual-systems
  4. https://link.springer.com/article/10.1007/s00221-017-4917-4
  5. https://pmc.ncbi.nlm.nih.gov/articles/PMC8807798/
  6. https://www.sciencedirect.com/science/article/abs/pii/S0010945217304094
  7. https://doi.org/10.1016/0166-2236(92)90344-8
  8. https://academic.oup.com/book/4415
  9. https://doi.org/10.1093/cercor/4.5.470
  10. https://www.science.org/doi/10.1126/science.163.3870.895
  11. https://link.springer.com/article/10.1007/s00259-018-4165-2
  12. https://www.nature.com/articles/349154a0
  13. https://www.annualreviews.org/content/journals/10.1146/annurev.neuro.20.1.337
  14. https://www.jneurosci.org/content/17/23/9280
  15. https://www.cell.com/neuron/fulltext/S0896-6273(12)00274-7
  16. https://www.jneurosci.org/content/39/24/4760
  17. https://www.jneurosci.org/content/17/11/4302
  18. https://www.sciencedirect.com/science/article/pii/S1053811904000576
  19. https://www.nature.com/articles/33402
  20. https://www.nature.com/articles/s41598-024-80546-4
  21. https://www.sciencedirect.com/science/article/abs/pii/S002839321730249X
  22. https://www.sciencedirect.com/science/article/pii/0166223692903448
  23. https://www.researchgate.net/publication/9014300_Two_Different_Streams_Form_the_Dorsal_Visual_System_Anatomy_and_Functions
  24. https://cris.unibo.it/bitstream/11585/585114/5/galletti2017.pdf
  25. https://www.vis.caltech.edu/documents/9417/coordinate_transformation.pdf
  26. https://pubmed.ncbi.nlm.nih.gov/9510384/
  27. https://pubmed.ncbi.nlm.nih.gov/15208633/
  28. https://www.jneurosci.org/content/29/39/12031
  29. https://pmc.ncbi.nlm.nih.gov/articles/PMC5444378/
  30. https://www.pnas.org/doi/10.1073/pnas.0805234105
  31. https://www.sciencedirect.com/science/article/abs/pii/S0010027703002282
  32. https://pubmed.ncbi.nlm.nih.gov/15110033/
  33. https://www.pnas.org/doi/10.1073/pnas.97.22.11800
  34. https://www.sciencedirect.com/topics/psychology/dorsal-stream
  35. https://pmc.ncbi.nlm.nih.gov/articles/PMC3021714/
  36. https://www.sciencedirect.com/science/article/pii/S0960982202011430
  37. https://pubmed.ncbi.nlm.nih.gov/12961051/
  38. https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0000424
  39. https://www.sciencedirect.com/science/article/abs/pii/S0167876005001091
  40. https://pubmed.ncbi.nlm.nih.gov/27318997/
  41. https://www.nature.com/articles/s41598-024-62098-9
  42. https://www.jneurosci.org/content/42/25/5070
  43. https://elifesciences.org/reviewed-preprints/94902
  44. https://doi.org/10.1038/349154a0
  45. https://pmc.ncbi.nlm.nih.gov/articles/PMC3613278/
  46. https://link.springer.com/article/10.1007/s12264-025-01371-x
  47. https://www.sciencedirect.com/science/article/pii/S0149763424001192
  48. https://www.nature.com/articles/s41562-022-01507-3
Add your contribution
Related Hubs
User Avatar
No comments yet.