Premotor theory of attention

• Dr. Giacomo Rizzolatti, Universita' Degli Studi di Parma, Italy

• Laila Craighero, University of Ferrara, Ferrara, Italy

Figure 1: Vertical meridian effect. Schematic drawing of the experimental set up used by Rizzolatti, Riggio, Dascola, Umiltà (1987) to test the presence of the Vertical meridian effect. The visual display comprised one fixation box and four boxes for stimulus presentation. The stimulus boxes were marked by an adjacent digit (1-4). The fixation box was always shown at the geometrical centre of the screen, whereas the position of the other boxes varied according to different blocks (horizontally in the upper or lower hemifield, vertically in the right or left hemifield). The cue used for directing attention was a digit (1-4), presented in the fixation box and indicating a probability of 80% for the target to appear in the relative box (valid trial) and 20% in one of the other possible boxes (invalid trials). Participants had to respond to target appearance by pressing a key with their right index finger. The importance of maintaining fixation was stressed and eye movements were monitored. The presence of the Vertical meridian effect was specifically tested by comparing reaction times to stimulus presentation in two locations that were at the same distance from the cued box (e.g., box 2) but in different hemifields. This comparison was significant and showed that reaction times were faster in the same (e.g., stimulus in box 1) than in the opposite hemifield (e.g., stimulus in box 3), indicating that passing from one hemifield to the other caused an additional delay of about 21 ms, regardless of which meridian had to be crossed.

Spatial attention is the capacity to improve the processing of sensory information coming from a specific part of the space surrounding the observer. Classically, spatial attention was thought of as a dedicated supramodal control mechanism, anatomically distinct from the circuits underlying sensorimotor processing (see Posner and Dehaene, 1994). In the late eighties Rizzolatti et al. (1987) challenged this view. On the basis of some behavioral experiments (see below) they argued that there is no need to postulate two control mechanisms, one for action and one for attention. According to them spatial attention does not result from a dedicated control mechanism, but derives from a weaker activation of the same frontal-parietal circuits that, in other conditions, determine motor behavior toward specific spatial locations. This theory, known as the “premotor theory of attention”, has received in these last years a tremendous support from electrophysiological and brain imaging studies and has been extended from spatial attention to attention directed to objects (Rizzolatti and Craighero, 1998).

Premotor theory of spatial attention

Psychological evidence in favor of the premotor theory of spatial attention

A first set of data that challenged the classical theory of attention came from behavioral experiments that used a variant of the Posner paradigm. In these experiments, unexpected imperative stimuli were located either in the same or in the opposite hemifield with respect to the attended location (see Figure 1). The results showed that when unexpected imperative stimuli were located in the hemifield contralateral to where attention was located, reaction times (RTs) were longer than when the imperative stimuli and attention were deployed in the same hemifield, even when the distance from the unattended stimuli and the cued location was the same (Rizzolatti et al., 1994).

This reaction time delay (the so called meridian effect) hardly can fit the notion that attention is a control system independent of basic anatomical and physiological circuits. In fact, there is no reason whatsoever why an anatomical landmark such as the meridian of the visual field could affect the function of a supramodal control mechanism. On the contrary, the meridian effect can be easily accounted for by assuming that attention derives from preparation to move the eyes towards the cued location (premotor theory of attention). When a cue indicates the location of the imperative stimulus, an eye movement program is prepared toward the expected location. This program specifies the direction and the amplitude of the saccade. If a target does not appear in the cued location, a new eye movement program has to be prepared. This requires first a selection of the saccade direction and then the setting of the saccade amplitude. Thus, changes in saccade direction require a radical modification in the oculomotor program, while changes in saccade amplitude imply only a readjustment of an existing program.

The same process occurs with attention deployment in the absence of eye movements. When an unexpected imperative stimulus appears in the same hemifield where the attention is deployed, only the amplitude of attention program has to be modified. In contrast, when the unexpected imperative stimulus occurs in the hemifield opposite to the cued one, the direction of attention has to be modified. Because the direction change involves the construction of a new ocular motor program, this process is more time-consuming.

Figure 2: Deviations of ocular saccades when attention is allocated to a peripheral point. The figure reports an example of the experimental paradigm used by Sheliga, Riggio, and Rizzolatti (1995) to support the premotor theory of attention. The basic assumption was that if visuospatial attention involves an activation of oculomotor circuits, then this activation should influence an overt oculomotor response. In contrast, if attention is not related to oculomotion, there is no reason why such an influence should occur. The basic experimental situation was similar to that used in experiments on meridian effect (see Figure 1). Participants were instructed to fixate a central cross and to pay attention to one of the possible stimulus locations according to the cue. The major difference was that the measured variable was not a manual response but a vertical saccade directed to a box located below or above the fixation point. Specifically, the visual display comprised three filled and two empty boxes. The three filled boxes were arranged horizontally. The two empty boxes were located above and below the central filled box. The central box was the fixation box. The cue was a thin line attached to this box, indicating the left box if pointing left, the right box if pointing right, the central box if pointed up. The stimulus consisted in a white vertical line across one of the filled boxes. In response to stimulus presentation, participants had to make a vertical saccade directed to one of the empty boxes, according to instructions. The figure shows all saccadic responses from one representative participant. Left side. Saccadic trajectories in response to imperative stimulus in the left box (upper panel), in the central box (middle panel), in the right box (lower panel). Right side. Mean horizontal deviation of saccades with imperative stimulus presented in the left box (upper panel) or in the right box (lower panel) after point-by-point subtraction of the mean horizontal deviation of saccades with the imperative stimulus presented in the central box (abscissa: horizontal deviation; ordinate: time). Main results showed that saccade trajectory deviated contralaterally to stimulus location, in particular the strength of the deviation was greater when trials were valid, that is when active attention was on the locus of imperative stimulus presentation. Finally, the deviation was present also when the imperative stimulus consisted not in a visual one, but in a computer-generated sound presented centrally, demonstrating that the deviation of response saccades doesn’t depend from the location of visual stimulus presentation.

Strong evidence in favor of the premotor theory of attention came from a series of experiments that used vertical saccades as the measured variables, and visual or acoustic stimuli as attention cues. The results showed that when participants paid attention to a given spatial location, the trajectory of a saccade triggered by an imperative stimulus deviated contralateral to the attention site ( Figure 2, Sheliga et al., 1995). This finding indicates that allocating attention to a given position necessarily activates the eye movement system, even if no ocular movement is required. Experiments showing that the discrimination accuracy is higher when the discrimination target is presented at the saccade target than when it is presented at adjacent positions (Deubel and Schneider, 1996) are in line with this conclusion.

Finally, compelling evidence for a causal relation between eye movements and attention was provided by experiments in which individuals could not perform a saccade towards the cued position either because a constrained extreme eye deviation or as a consequence of a peripheral palsy. The rational underlying these experiments was the following. If the involvement of the oculomotor system during spatial attention tasks reflects a causal relationship between eye movement programming and attention, the modifications in eye movement abilities should be paralleled by a modification in the ability to orient visuospatial attention, while, if not, the eye position in the orbits should be irrelevant. The results showed that when the eyes cannot move towards a certain location also the attention cannot move (Craighero et al., 2001; Craighero et al., 2004).

Neuroimaging evidence

A strong support for the premotor theory of attention came from neuroimaging studies. These studies clearly showed that visuospatial attention and eye movements share the same cortical neuronal network (Corbetta et al., 1998; Nobre et al., 2000). No system of distinct cortical areas was activated exclusively by covert attention or by a saccade task ( Figure 3). The network found in those studies includes the putative human homologue of monkey frontal eye fields (FEF) and the lateral intraparietal area (LIP), both areas known to be involved in voluntary control of eye movements.

Figure 3: Brain regions active during overt and covert shifts of attention. Comparison of responses to overt (left panel) and covert (right panel) shifts of attention in a single subject, reported by Beauchamp, Petit, Ellmore, Ingeholm, Haxby (Neuroimage. 2001, 14:310-321). Subjects were required to made either overt attentional shifts or covert attentional shifts. Both overt and covert shifts of visuospatial attention induced activations in the precentral sulcus, intraparietal sulcus, and lateral occipital cortex that were of greater amplitude for overt than during covert shifting, reflecting additional activity associated with saccade execution. These results confirmed that overt and covert attentional shifts are subserved by the same network of areas. This network includes the putative human homologue of monkey frontal eye fields (FEF) that is known to be involved in voluntary control of eye movements. Color scale indicates significance of activation. PreCS, precentral sulcus; IPS, intraparietal sulcus.

A recent event-related fMRI experiment on congenitally blind individuals (Garg et al., 2007) supported the notion that eye movement and attention planning cannot be separated. In a covert orienting task, with endogenous verbal cues and lateralized auditory targets, these authors found a robust stimulus-locked FEF activation in congenitally blind individuals, similar to that observed in sighted controls with eyes closed.

Neurophysiological evidence

Further support to the premotor theory of attention came from physiological studies in monkeys. Kustov and Robinson (1996) recorded neuron activity from the monkey superior colliculus, a center crucially involved in eye movements. The results showed an increase in the excitability of this structure when a monkey paid attention to a given location in space. Particularly striking was the observation that the collicular excitation also changed when the monkey was instructed to make a manual response and to keep the eyes still after imperative stimulus presentation.

Figure 4: Effect of underthreshold FEF stimulation on psychophysical performance. A) Individual saccade vectors obtained with suprathreshold stimulation at an FEF site in the monkey, illustrating how the motor field was mapped. Vector traces show eight saccades evoked in eight trials at a current of 38 mA. B) Illustration of the visual display viewed by the monkey. The monkey was trained to fixate on the central fixation spot and attend to a peripheral target that could transiently dim at a random time during the trial, in presence of visual distractors randomly flashed throughout the display. C) Staircase data and luminance change threshold estimates (% contrast) obtained with (black circles) and without (white circles) microstimulation (stimulation onset asynchrony=175 ms). Results showed that microstimulation improved monkey’s performance when, but only when, the object to be attended was positioned in the space represented by the cortical stimulation site (Moore & Fallah, 2001).

In a very brilliant electrophysiological experiment Moore and Fallah (2001) reported that it is possible to enhance spatial perception by altering oculomotor signals within the brain. The authors trained two monkeys to make manual responses at the detection of a transient dimming of a peripheral visual target and tested the effects of FEF microstimulation on monkeys’ performance. The results showed that subthreshold stimulation of a particular subregion of FEF determined a decrease in the psychophysical threshold for stimulus detection, but only when the target stimulus was positioned in the motor field corresponding to the stimulation point. This finding provided evidence of a direct effect of eye movement control on the allocation of spatial attention ( Figure 4).

Subsequently, Moore and Armstrong (2003) also showed that a subthreshold microstimulation of the FEF enhanced visual responses in V4 neurons located at corresponding spatial locations. Similar results have been also obtained by Ekstrom et al. (2008) by combining fMRI and chronic electrical microstimulation in awake, behaving monkeys. Stimulation of subregions in the FEF determined a strong activation of higher-order visual areas, as those obtained during shifts in attention. These findings suggest that the gain of visual responses in extrastriate cortex is modulated by the same activity that elicits a saccade to a particular location. This provides evidence for a mechanism modulating visual responses at the attended locations, when saccades are planned, but not executed. The results establishing a causal effect of FEF stimulation on visual cortex have been recently confirmed in humans by Ruff and colleagues (Ruff et al., 2006). In an experiment combining fMRI with transcranial magnetic stimulation (TMS) they showed that stimulating human FEF produces systematic effects on fMRI signal in early human visual cortex, including even area V1.

Spatial attention related to arm movements planning

In everyday life most of our actions in space are preceded by foveation and this gives the eye movement system a special central position in spatial attention. There are, however, some conditions in which we do not use, or do not use primarily, eye movements to select stimuli in space. In these cases spatial attention should depend on circuits other than those related to eye movements. Probably the best documented evidence in favor of spatial attention not related to eye movements is that deriving from experiments conducted by Tipper et al. (1992). They studied, in normal participants, the effect of an irrelevant stimulus (distractor) located in or out of the arm trajectory necessary to execute a pointing response. The result showed that an interference effect was present only when the distractor was located in the trajectory necessary to execute a pointing response. Control experiments suggested that the effect was not due to a purely visual representation of the stimuli or to spatial attention related to eye movements. Rather, the organization of the arm-hand movement determined a change in the attentional relevance of stimuli close to the hand or far from it.

Similar results are those by Eimer and colleagues on covert manual movement preparation (Eimer et al. 2005). The authors recorded ERPs in normal volunteers during the interval between a visual cue and a subsequent visual Go/Nogo signal. During this interval the volunteers were instructed to prepare to lift their left or right index finger. Results showed that somatosensory ERP components were enhanced when task-irrelevant tactile probes were delivered during response preparation to the hand involved in an anticipated response, indicating that covert manual movement preparation influences somatosensory processing.

Premotor theory of object-related attention

Introduction

Recently, a series of experiments tested whether the premotor theory of attention could be extended from orienting of attention to spatial locations to orienting of attention to graspable objects. This hypothesis came from neurophysiological studies of the monkey parietal (anterior intraparietal area, AIP) and premotor cortex (F5). These studies showed that many neurons located in these areas selectively discharge during the execution of object grasping and some also during object observation. The visual responses of the latter neurons were only present when there was a congruence between the intrinsic properties of the object (size and shape) and the coded grip (precision grip, whole hand prehension) (Rizzolatti and Luppino, 2001). These findings indicate that every time a graspable object is observed, these visuo-motor neurons are activated eliciting a potential specific motor act. On the basis of these observations, it is possible to draw a parallelism between spatial attention and object-related attention: as in the case of spatial attention eye movement preparation selects a given spatial location, the preparation of a grasping movement selects an object with specific intrinsic characteristics.

Psychophysical evidence

Tucker and Ellis (1998) provide evidence that a potential specific motor act is evoked every time a graspable object is presented. Normal human volunteers were presented with photographs of common graspable objects. The volunteers had to respond with their left or right hand. The results showed that the reaction times were faster when the response was executed with the hand most suited to grasp the presented object. Very recently, Adamo and Ferber (2009), using the attentional blink paradigm and event-related potentials, confirmed these findings demonstrating that the presentation of a tool leads to attentional enhancement towards subsequently presented objects that are consistent with the action afforded by the tool.

In a recent experiment it has been tested if the shape of the to-be-grasped object influences observer’s prediction about the way in which the agent will act onto that object (Craighero et al. 2008). To this purpose volunteers were requested to detect the instant at which the demonstrator’s hand touched the object during a grasping and lifting action. Two types of grasping were presented, differing for the type of fingers opposition space: in one case the grasping was the one more commonly used to grasp the presented object, in the other case it was a less appropriate one. In this way a conflict was present when the action evoked by object observation didn’t coincide with that executed by the experimenter.

Results showed that the response times were shorter for ‘suitable’ grasping trials than for ‘not suitable’ ones, indicating that the observed action is recognized faster when the corresponding action representation is already activated by the vision of the to-be-grasped object. This facilitation reflects a modification in the perceptual salience of others’ action, thus linking this effect to classical attentional phenomena.

Figure 5: Effects of motor preparation on object detection. A) Experimental set up used by Craighero, Fadiga, Rizzolatti, and Umiltà, (1999) to verify the influence that the preparation of a grasping movement determines on the detection of objects sharing or not the same intrinsic properties of the to-be-grasped object. Participants were seated in front of the computer screen with their head positioned on a chin rest and pressing a switch with their right hand shaped in a pinch position. B) Schematic drawing of the experimental design. The go signal consisted of one of two pictures: a rectangle rotated 45° clockwise or a rectangle rotated 45° counterclockwise. The response consisted of grasping, as fast as possible, a red plastic bar inserted inside a rectangular hole hollowed in a white plastic disk and glued to it, located in front of the switch (distance: 12 cm), out of the participant's sight. The orientation of the bar could be either clockwise or counterclockwise oriented. There were two experimental conditions. In the congruent condition, the go signal was a rectangle the orientation of which matched the orientation of the bar to be grasped. In the incongruent condition, the go signal was a rectangle the orientation of which did not match the orientation of the bar to be grasped. C) Main results indicated that reaction times (the time between the go signal and the releasing of the switch) were faster for congruent condition than for incongruent condition, suggesting that motor preparation determines a faster processing of object intrinsic properties shared by both the to-be-grasped object and the visual one: when one is ready to grasp an object, she is faster in recognizing it. (CON, congruent condition; INC, incongruent condition).

In line with these findings is the observation that the preparation of a grasping movement to a given object determines the allocation of attention to the intrinsic characteristic of that object. Craighero et al., (1999) asked volunteers to be ready to grasp a bar and then to grasp it as fast as possible on the presentation of a visual rectangle differently oriented ( Figure 5). The results showed that grasping reaction times to rectangles congruently oriented were faster than reaction times to those incongruently oriented. These data indicate that the preparation to act on an object produces a faster visual processing of the stimuli congruent with that object.

Conclusions

Premotor theory of attention has been originally advanced on the basis of indirect evidence plus anatomical and physiological considerations on the organization of the nervous circuits involved in sensorimotor transformation. In these last years a large number of evidence coming from neurophysiological studies on monkeys, and behavioral and brain imaging studies in humans, clearly demonstrated the validity of the main tenet of the premotor theory of attention, that is that attention derives from the activity of the sensorimotor circuits, rather than from a specific center devoted to attention. Furthermore, while the initial version of the premotor theory of attention concerned only those aspects of attention that are linked to the oculomotor system, recent evidence enlarged the scope of the theory including attention effects deriving from motor programs related to reaching and grasping movements. This enlargement provides a unitary explanation of a variety of visual-related attentional phenomena.

References

• Adamo, M. and Ferber, S. (2009). A picture says more than a thousand words: behavioural and ERP evidence for attentional enhancements due to action affordances. Neuropsychologia. 47:1600-8.
• Corbetta, M., Akbudak, E., Conturo, T.E., Snyder, A.Z., Ollinger, J.M., Drury, H.A., Linenweber, M.R., Petersen, S.E., Raichle, M.E., Van Essen, D.C. and Shulman, G.L. (1998). A common network of functional areas for attention and eye movements. Neuron. 21: 761-73.
• Craighero, L., Bonetti F., Massarenti, L., Canto, R., Fabbri Destro, M. and Fadiga, L. (2008). Temporal prediction of touch instant during observation of human and robot grasping. Brain Res Bull. 75:770-4.
• Craighero, L., Carta, A. and Fadiga, L. (2001). Peripheral oculomotor palsy affects orienting of visuospatial attention. NeuroReport 12: 3283-3286.
• Craighero, L., Fadiga, L., Rizzolatti, G. and Umiltà, C. (1999). Action for perception: a motor-visual attentional effect. J. Exp. Psychol. Hum. Percept. Perform. 25: 1673-1692.
• Craighero, L., Nascimben, M. and Fadiga, L. (2004). Eye position affects orienting of visuospatial attention. Curr Biol. 14:331-3.
• Deubel, H. and Schneider, W.X. (1996). Saccade target selection and object recognition: evidence for a common attentional mechanism. Vision Res. 36:1827-37.
• Eimer, M., Forster, B., Van Velzen, J. and Prabhu, G. (2005). Covert manual response preparation triggers attentional shifts: ERP evidence for the premotor theory of attention. Neuropsychologia.43:957-66.
• Ekstrom, L.B., Roelfsema, P.R., Arsenault, J.T., Bonmassar, G. and Vanduffel, W. (2008). Bottom-up dependent gating of frontal signals in early visual cortex. Science. 321:414-7.
• Garg, A., Schwartz, D. and Stevens, A.A. (2007). Orienting auditory spatial attention engages frontal eye fields and medial occipital cortex in congenitally blind humans. Neuropsychologia. 45:2307-21.
• Gherri, E. and Eimer, M. (2008). Links between eye movement preparation and the attentional processing of tactile events: an event-related brain potential study. Clin. Neurophysiol. 119(11):2587-2597.
• Gherri, E., Driver, J. and Eimer, M. (2008). Eye movement preparation causes spatially-specific modulation of auditory processing: new evidence from event-related brain potentials. Brain Res. 1224:88-101.
• Juravle, G. and Deubel, H. (2009). Action preparation enhances the processing of tactile targets. Exp Brain Res. 198(2-3):301-311.
• Kustov, A.A. and Robinson, D.L. (1996). Shared neural control of attentional shifts and eye movements. Nature 384: 74-7.
• Moore, T. and Armstrong, K.M. (2003). Selective gating of visual signals by microstimulation of frontal cortex. Nature 421: 370–373.
• Moore, T. and Fallah, M. (2001). Control of eye movements and spatial attention. Proc. Natl. Acad. Sci. USA. 98: 1273-6.
• Nobre, A.C., Gitelman, D.R., Dias, E.C. and Mesulam, M.M. (2000). Covert visual spatial orienting and saccades: overlapping neural systems. Neuroimage 11: 210-6.
• Posner, M.I. and Dehaene, S. (1994). Attentional networks. Trends neurosci. 17: 75-79.
• Rizzolatti, G. and Craighero, L. (1998). Spatial attention: Mechanisms and theories. In M. Sabourin, F. Craik, & M. Robert (Eds.), Advances in psychological science: Vol.2. Biological and cognitive aspects (pp. 171-198). East Sussex, England: Psychology Press.
• Rizzolatti, G. and Luppino, G. (2001). The cortical motor system. Neuron. 31:889-901.
• Rizzolatti, G., Riggio, L. and Sheliga, B.M. (1994). Space and selective attention. In C. Umiltà & M. Moscovitch (Eds.), Attention and Performance XV (pp. 231-265). Cambridge, MA: MIT Press.
• Rizzolatti, G., Riggio, L., Dascola, I. and Umiltá C. (1987). Reorienting attention across the horizontal and vertical meridians: evidence in favor of a premotor theory of attention. Neuropsychologia 25: 31-40.
• Rorden, C. and Driver, J. (1999). Does auditory attention shift in the direction of an upcoming saccade? Neuropsychologia 37(3):357-377.
• Rorden, C., Greene, K., Sasine, G. and Baylis, G. (2002). Enhanced tactile performance at the destination of an upcoming saccade. Curr Biol. 12(16):1429-1434.
• Ruff, C.C., Blankenburg, F., Bjoertomt, O., Bestmann, S., Freeman, E., Haynes, J.D., Rees, G., Josephs, O., Deichmann, R. and Driver, J. (2006). Concurrent TMS-fMRI and psychophysics reveal frontal influences on human retinotopic visual cortex. Curr Biol. 16:1479-88.
• Sheliga, B.M., Riggio, L. and Rizzolatti, G. (1995). Spatial attention and eye movements. Exp. Brain Res. 98: 507-522.
• Tipper, S.P., Lortie, C. and Baylis, G.C. (1992). Selective reaching: Evidence for action-centered attention. J. Exp. Psychol. Hum. Percept. Perform. 23: 823-844.
• Tucker, M. and Ellis, R. (1998). On the relations between seen objects and components of potential actions. J. Exp. Psychol. Hum. Percept. Perform. 24: 830-846.

Internal references

• Lawrence M. Ward (2008) Attention. Scholarpedia, 3(10):1538.

• Kimron L. Shapiro, Jane Raymond, Karen Arnell (2009) Attentional blink. Scholarpedia, 4(6):3320.

• Valentino Braitenberg (2007) Brain. Scholarpedia, 2(11):2918.

• James Meiss (2007) Dynamical systems. Scholarpedia, 2(2):1629.

• Keith Rayner and Monica Castelhano (2007) Eye movements. Scholarpedia, 2(10):3649.

• William D. Penny and Karl J. Friston (2007) Functional imaging. Scholarpedia, 2(5):1478.

• Seiji Ogawa and Yul-Wan Sung (2007) Functional magnetic resonance imaging. Scholarpedia, 2(10):3105.

• Rodolfo Llinas (2008) Neuron. Scholarpedia, 3(8):1490.

• Anthony T. Barker and Ian Freeston (2007) Transcranial magnetic stimulation. Scholarpedia, 2(10):2936.

Further reading

• None

External links

• Giacomo Rizzolatti's website
• Laila Craighero's website

See also

Attention, Attentional blink, Brain

Sponsored by: Eugene M. Izhikevich, Editor-in-Chief of Scholarpedia, the peer-reviewed open-access encyclopedia

Sponsored by: Naotsugu Tsuchiya, California Institute of Technology, Pasadena, CA, USA

Reviewed by: Anonymous

Accepted on: 2009-11-16 16:31:42 GMT