During the past few decades, the development and refinement of neuroimaging techniques has facilitated significant progress toward the diagnosis and treatment of neurological disorders. Perhaps the greatest advances have occurred in the field of magnetic resonance imaging (MRI). The spatial resolution of MRI is now less than 1 mm, allowing researchers and clinicians to recognize minute structural abnormalities brought about by development, pathology, or injury. Although technologies such as computerized tomography (CT) and positron emission tomography (PET) have contributed to improved understanding and management of specific neurological disorders, much of the neuroimaging research in aphasia has relied on MRI technology.

In general, most neuroimaging research on aphasia recovery can be divided into two categories: studies of early, spontaneous recovery, including the acute and/or sub-acute phases of stroke; and studies of treated recovery, usually occurring in the chronic phase of stroke. Most studies targeting the early phases of spontaneous recovery have used structural- and perfusion-weighted MRI to examine how structural damage or compromised cerebral perfusion influences early aphasia progression. In the case of treated recovery from chronic aphasia, the bulk of research has used functional MRI (fMRI) to identify functional brain changes (for a brief explanation of the different MRI techniques, see images throughout the article).

Neuroimaging of Early Recovery

Much of the neuroimaging work related to spontaneous recovery from aphasia in the acute phase of stroke has been spearheaded by Argye Hillis, a speech-language pathologist and practicing neurologist. In a seminal study, Hillis and colleagues (2000) examined 40 patients with acute stroke and discovered that the volume of hypoperfusion (decreased blood flow at the capillary level) was a stronger predictor of neurological deficit (aphasia or hemi-spatial neglect) than the volume of lesion size seen with structural MRI using diffusion-weighted imaging (DWI). This study revealed that compromised perfusion beyond the structural brain damage was common in acute stroke, suggesting that an increase in cerebral perfusion might spur at least partial recovery from the neurological deficit. Medical treatment to improve perfusion was administered to a subset of the patients, with a favorable outcome. Those patients who experienced reperfusion of the previously hypoperfused brain areas also demonstrated significant resolution of either aphasia or hemi-spatial neglect.

In a more recent study, Hillis and her team (2006) examined a large group of patients who underwent perfusion- and diffusion-weighted MRI scans and language testing at bedside within hours of stroke onset. A subset of the patients received treatment to restore blood flow to areas that were hypoperfused and all were re-examined four to five days later using the same MRI scans and language testing. What was particularly revealing in this study was the strong relationship between anomia recovery and reperfusion to portions of Broca's area, Wernicke's area, and, in particular, the left posterior, middle, and inferior temporal lobes. Although this study targeted only patients who received treatment to boost cerebral perfusion, it seems reasonable to expect that some of the early spontaneous recovery seen in aphasia is associated with improved cerebral perfusion. Naturally, this result does not discount other recovery processes such as lessening of edema and changes in glutamatergic activity. Nevertheless, these findings suggest that significant spontaneous aphasia recovery in the first few days of stroke is supported, at least partially, by changes in cerebral perfusion rather than in structural reorganization.

Stressing patients' need for healing and adaptation, some have argued for a primary counseling role for SLPs in acute aphasia management (Fridriksson & Holland, 2001; Holland & Fridriksson, 2001). Based on their unique expertise with aphasia, SLPs are probably better suited than other health care professionals to counsel both patients and family members about the dynamic aspects of spontaneous recovery of communicative function. Given that macro changes in brain physiology are probably not positively influenced by traditional aphasia treatment in the acute phase of stroke, it seems that arguing for patient counseling and postponing aphasia treatment until the sub-acute or chronic phase of stroke also can be motivated by neurophysiological factors. That is, the existence of brain edema and hypoperfusion probably make for a sub-optimal brain status for taking on necessary cortical changes needed to support recovery associated with behavioral treatment. Thus, waiting to start aphasia treatment until early neurophysiological changes have run their course may give patients time to adjust to the negative psycho-social aspects of aphasia and prepare for what is often a long road to recovery.

Aphasia Recovery Patterns 

The pattern of aphasia recovery varies significantly among patients (Kertesz & McCabe, 1977) and many patients experience some spontaneous recovery well after what is traditionally defined as the chronic phase of stroke (Lendrem & Lincoln, 1985). To examine the changes in brain activation as patients progress from the acute to the chronic phase of stroke, a group of researchers led by Dorothee Saur (2006) scanned 14 patients with aphasia using fMRI in the acute, sub-acute, and chronic phases of stroke recovery. fMRI during the acute stage (mean time post-stroke = 1.8 days) revealed very depressed overall brain activation, especially in the damaged left hemisphere. At an average of 12 days post-stroke, an overall increase in brain activation was revealed, with particularly strong activation in the right hemisphere Broca's area homologue, which correlated with overall aphasia recovery. A final fMRI scan at approximately 320 days post-stroke revealed the greatest activation in the left hemisphere. This "reactivation" of the left hemisphere was associated with aphasia recovery. Overall, this study emphasizes the dynamic reorganization of brain function that occurs following stroke. There is not yet evidence to suggest that aphasia treatment influences the progressive dynamics of brain reorganization during the early phases of recovery. It would seem crucial to understand whether favorable brain plasticity during early recovery can be positively influenced or whether the same use of resources might be better used once spontaneous recovery processes have run their course.

As with studies of early recovery, relatively few studies have tackled the relationship between brain activation and aphasia treatment outcome. Much of this research has relied on single-subject designs, making it difficult to generalize the findings beyond the patients included in each study. Several group studies, however, have examined treatment-related brain changes as predictors of outcome. In one such study, Maria Richter and colleagues (2008) used fMRI to examine the relationship between brain activation and the outcome of constraint-induced language therapy in 16 participants with chronic, non-fluent aphasia following left-hemisphere stroke. In short, their study revealed localized decrease in cortical activation of specific areas of the right hemisphere as a predictor of positive aphasia treatment outcome. Unfortunately, this study did not examine possible treatment-related changes in the left hemisphere, making it impossible to infer whether the decreased right hemisphere activation was related to greater reliance on the left hemisphere.

Treatment-Induced Neuroplasticity

Interestingly, this work ties in nicely with that of Margaret Naeser and colleagues (2005), who used transcranial magnetic stimulation (TMS) to decrease activity in the right homologue of Broca's area and subsequently improve the naming abilities of persons with chronic, non-fluent aphasia. Their work suggests that brain stimulation that inhibits the right hemisphere may improve language functioning in aphasia. Again, it is not clear whether such active inhibition increases reliance on preserved brain regions of the left hemisphere. In a 2010 study, our group examined brain activation associated with picture-naming in persons with chronic aphasia and found that increased reliance on the preserved left hemisphere was strongly related to decreased anomia severity (Fridriksson et al., 2010). More specifically, greater left-hemisphere activation was associated with improved picture-naming by participants with chronic aphasia than by normal control participants. Although this research did not tackle recovery per se, it certainly complements other studies that have shown an association between decreased right hemisphere activation and improved language processing. That is, it seems that the extent of language-related modulation of the left hemisphere predicts decreased aphasia severity. This observation also is supported by other studies that have yielded similar findings (Cornelissen et al., 2003; Meinzer et al., 2008; Postman-Caucheteux et al., 2010; Rosen et al., 2000).

Although this research suggests a crucial role of preserved left-hemisphere regions in aphasia recovery, it is important to point out that the dynamics of brain organization probably vary widely among individuals and that, in some cases, right-hemisphere recruitment may support aphasia recovery. Two recent studies are particularly revealing in this regard. In the first study, Bruce Crosson and colleagues (2009) examined changes in brain activation associated with aphasia treatment with the aim of shifting brain activity from the left to the right hemisphere. Of particular interest was the novelty of their aphasia treatment approach, in which picture-naming attempts were paired with productions of specific movements (opening a box and pressing a button) by the non-dominant, left hand. This pairing was hypothesized to promote greater reliance on the right hemisphere for picture-naming. This study, which included five participants with chronic aphasia, revealed that the four participants whose language processing improved also experienced increased right laterality of brain activation (i.e., greater right- compared to left-hemisphere activation as measured with fMRI). The remaining participant did not respond to treatment and showed greater left-hemisphere activity upon treatment completion.

The second study of interest also used a treatment approach, Melodic Intonation Therapy (MIT), designed to target the intact right hemisphere in persons with aphasia (Albert et al., 1973). MIT was designed for individuals with non-fluent aphasia and assumes that pairing speech with melody (i.e., singing) recruits the right hemisphere and increases verbal output. Gottfried Schlaug and colleagues (2009) examined the density of white-matter fibers in the right hemisphere in six participants with chronic, non-fluent aphasia before and after receiving MIT. Using diffusion tensor imaging (DTI), an MRI technique that allows for tracking of the white-matter fibers in the brain, Schlaug et al. found that the density of the white-matter tracts in the right hemisphere had increased among the participants upon completion of MIT. As in the case of Crosson et al. (2009), these results suggest that specific aphasia treatment approaches can alter both function and structure of the right hemisphere with corresponding positive behavioral outcomes.

These studies also highlight the possibility that different aphasia treatment approaches target different parts of the brain areas that survived the stroke. For example, in cases of larger left-hemisphere lesions, treatments shown to modulate the right hemisphere may be more appropriate than approaches in which success relies on recruitment of the left hemisphere. Similarly, treatments that promote left-hemisphere plasticity may be more appropriate in the case of smaller lesions in which positive treatment outcome relies on re-establishment of the damaged cortical language network in the left hemisphere. The selection of a specific treatment approach for a given patient with aphasia relies on clinical intuition as well as on scientific evidence. Clinical intuition based on extensive experience represents an important aspect of aphasia management. However, if the extent and location of brain damage relate to outcomes with different aphasia treatment approaches, then it is possible that SLPs may rely, in part, on neuroimaging data to improve treatment outcome.

Neurophysiological Dynamics of Stroke Recovery 

Although the evaluation of brain damage and brain plasticity may prove to be important for treatment selection, we certainly are not there yet. However, understanding areas of favorable brain plasticity is already being used to guide aphasia treatment research. In a recent study (Baker, Rorden, & Fridriksson, 2010), we used transcranial direct current stimulation (tDCS) in conjunction with a computerized aphasia treatment (see Fridriksson et al., 2009) to examine the added effect of non-invasive brain stimulation on anomia treatment outcome in persons with chronic aphasia caused by ischemic stroke. This double-blind study demonstrated that anodal tDCS applied to the left hemisphere significantly enhances the effect of anomia treatment in aphasia. To ensure that the electrical stimulation was applied to a brain area recruited for picture-naming and to prevent electrode placement over damaged tissue, both fMRI and structural MRI were used to guide stimulation. This method enabled the researcher to tailor the location of brain stimulation to the location of each participant's lesion location and brain activation. Although the application of brain stimulation to treat aphasia warrants far more scrutiny, the use of MRI-guided TMS and tDCS seems plausible as a future means to improve aphasia treatment outcome.

So far, the application of neuroimaging has revealed much about the physiology of early stroke recovery, including that of aphasia. Although it is not entirely clear how this informs SLPs' early treatment of aphasia, it is possible that a better understanding of the neurophysiological dynamics of stroke will allow us to manage our resources better during each phase of recovery and thereby maximize long-term aphasia recovery. For patients with chronic aphasia, neuroimaging has revealed that successful aphasia treatment, along with improving communication ability, does influence both brain function and brain structure. Although the obvious goal of aphasia treatment is to improve the patients' ability to communicate, understanding how aphasia treatment influences the brain may, in turn, improve the selection of the specific treatment approach.