Can Computers Read Emotions in Therapy? New Research Explores AI Analysis of Therapy Sessions
When people go to therapy, emotions are at the heart of the process. Patients share their feelings, work through difficult experiences, and hopefully develop healthier ways of managing their emotional lives. But measuring and tracking those emotions throughout treatment has always been challenging for both researchers and therapists.
A recent study explored whether artificial intelligence could help by automatically analyzing the emotional tone of therapy conversations. Researchers used computer algorithms to examine transcripts from therapy sessions, looking for signs of positive and negative emotions in patients' words. They then checked whether these AI detected emotions matched what patients and therapists actually reported feeling during those same sessions.
Reading Between the Lines
The research team analyzed 85 therapy sessions involving 35 adult patients receiving cognitive behavioral therapy at a German university clinic. They used a sophisticated AI system designed to detect sentiment in text, which had been trained on millions of social media posts and other written content in multiple languages including German.
The AI examined only what patients said during sessions, not therapist statements, focusing on detecting overall positive or negative emotional tone. Meanwhile, both patients and therapists filled out questionnaires after each session rating how the patient felt during that particular meeting, using measures of emotions like sadness, anxiety, contentment, and relaxation.
Promising Connections
The results showed encouraging signs that the AI analysis could capture meaningful emotional information. When the computer detected more positive sentiment in a patient's speech, both patients and therapists were more likely to report positive emotions in that same session. Similarly, when the AI picked up negative sentiment, human ratings of negative emotions tended to be higher.
These connections held up even when researchers used advanced statistical methods to account for the fact that the same patients appeared across multiple sessions and were treated by different therapists. The correlations weren't perfect, but they were substantial enough to suggest the AI was detecting real emotional patterns.
The researchers also found that AI detected sentiments connected to important therapy processes. When patients showed more positive sentiment, they were more likely to report having good coping experiences and feeling like they were mastering challenges during therapy. Negative sentiment was associated with sessions where patients engaged more deeply with difficult emotional material.
Perhaps most importantly, changes in sentiment over time predicted therapy outcomes. Patients whose speech became more positive throughout treatment showed better results on measures of anxiety, depression, and overall psychological functioning when therapy ended. Those whose sentiment became more negative tended to have worse outcomes.
A New Tool in the Toolkit
This research suggests that sentiment analysis could become a valuable addition to existing ways of monitoring progress in therapy. Currently, therapists rely primarily on patient self reports, their own observations, and standardized questionnaires to track how someone is doing. Computer analysis of therapy conversations could provide another perspective, potentially catching emotional patterns that humans might miss.
The approach could be particularly useful because it provides an objective measure that doesn't rely on patients remembering how they felt or being willing to accurately report their emotions. Some people struggle to identify or communicate their emotional states, and others might feel pressured to present themselves in a certain way to their therapist.
If this technology continues to develop, it could help therapists in several ways. They might receive feedback about emotional patterns in their sessions, helping them notice when a patient is struggling even if that person doesn't explicitly say so. The analysis could also flag sessions with sudden increases in negative emotion, potentially alerting therapists to times when extra support might be needed.
For therapy training and supervision, sentiment analysis could help new therapists learn to recognize emotional dynamics in their sessions. Supervisors could review transcripts with trainees, using the AI analysis to discuss how emotional tone shifted throughout a session and what that might mean for treatment.
The technology could also advance therapy research by making it easier to study emotional processes across large numbers of sessions. Currently, analyzing emotions in therapy often requires trained human raters, which is time consuming and expensive. Automated analysis could help researchers study emotional patterns on a much larger scale.
The researchers acknowledge several important limitations to their work. The study was relatively small, involving only 35 patients, and was exploratory in nature. The AI system they used wasn't specifically designed for therapy conversations, so a system trained specifically on therapy data might perform better.
There were also some discrepancies between what the AI detected and what therapists reported about patient emotions. This could mean the AI was missing something important, or it might suggest that therapists sometimes have difficulty accurately reading their patients' emotional states. More research is needed to understand these differences.
The study also required therapy sessions to be transcribed, which is currently time consuming and expensive. For this technology to be practical in real therapy settings, more efficient transcription methods would need to be developed.
The Human Element Remains Central
It's important to note that this research doesn't suggest computers should replace human judgment in therapy. Rather, sentiment analysis might serve as an additional source of information to complement therapists' clinical expertise and patients' self reports.
The researchers emphasize that their AI system only captures basic emotional tone, not the complex nuances of human emotion and experience that skilled therapists are trained to recognize and respond to. The goal is to augment human understanding, not replace it.
Future research in this area might combine sentiment analysis with other automated approaches, such as analyzing facial expressions, voice tone, or physiological measures like heart rate. This multimodal approach could provide an even richer picture of emotional processes during therapy.
Researchers are also working on developing AI systems specifically trained on therapy conversations, which might be more accurate than general purpose sentiment analysis tools. As automatic transcription technology improves, it may become more feasible to implement these approaches in routine clinical practice.
The study represents an early step toward understanding how artificial intelligence might support emotional assessment in therapy. While the technology isn't ready for widespread clinical use, the research suggests it could eventually become a valuable tool for therapists, researchers, and patients working to understand and improve emotional well being.
As mental health care continues to evolve, tools like sentiment analysis might help make therapy more responsive to patients' moment to moment emotional experiences, potentially improving outcomes for people seeking help with psychological difficulties.
Eberhardt, S. T., Schaffrath, J., Moggia, D., Schwartz, B., Jaehde, M., Rubel, J. A., ... & Lutz, W. (2025). Decoding emotions: Exploring the validity of sentiment analysis in psychotherapy. Psychotherapy Research, 35(2), 174-189.
Imagine if doctors could predict which type of therapy would work best for you before you even start treatment. A comprehensive study from Finland suggests this might soon be possible, using artificial intelligence to match patients with their optimal therapeutic approach.
Researchers analyzed data from over 2,000 adults who received short term psychotherapy through the Finnish healthcare system. They used advanced machine learning techniques to determine whether matching people to specific therapy types based on their characteristics could improve treatment outcomes compared to the current system where therapy assignment is often based on availability or general preferences.
Four Different Paths to Healing
The study examined four main types of therapy commonly used in Finland. Solution-focused therapy emphasizes building on existing strengths and working toward future goals. Cognitive-behavioural therapy focuses on changing unhelpful thought patterns and behaviours. Psychodynamic therapy examines how past experiences and unconscious processes influence current issues. Integrative and cognitive analytic therapies combine elements from various approaches, adapting flexibly to meet the individual needs of each patient.
Traditionally, research has suggested that these different therapy approaches produce roughly equal results on average, a finding known as the "dodo bird verdict" after the character in Alice in Wonderland who declared that "everyone has won and all must have prizes." However, this new research suggests that while therapies may be equally effective on average, specific individuals might benefit much more from one approach than another.
The Power of Personalized Matching
The researchers used a sophisticated statistical approach called targeted learning, which combines multiple machine learning algorithms to make predictions while accounting for the complex factors that influence both treatment assignment and outcomes. They fed the system information about patients' demographics, symptoms, medical history, social support, and other characteristics to predict which therapy would work best for each person.
When the algorithm assigned each patient to their predicted optimal therapy, treatment outcomes improved by about 0.3 standard deviations across multiple measures of mental health and functioning. To put this in perspective, this improvement would take someone from experiencing moderate benefits to experiencing good benefits from therapy.
This improvement is particularly meaningful because it's comparable to the difference between receiving therapy and receiving no treatment at all in many studies. In other words, getting the right type of therapy might be almost as important as getting therapy in the first place.
Different Goals, Different Recommendations
Interestingly, the optimal therapy recommendation for each person varied depending on what outcome the researchers were trying to maximize. Some patients might be predicted to respond best to one therapy for reducing depression symptoms, but a different treatment for improving their ability to function at work or in relationships.
Only about one-third of patients received the same therapy recommendation across all outcome measures, highlighting that the "best" therapy depends partly on what you're hoping to achieve. This suggests that clear goal setting at the beginning of treatment could be crucial for optimal therapy matching.
Who Benefits from What
The analysis revealed patterns indicating which types of patients were predicted to benefit from specific therapies. Patients recommended for solution focused therapy tended to have fewer social connections and support systems. Those assigned to psychodynamic therapy were typically younger, reported higher alcohol consumption, more loneliness, but fewer anxiety symptoms.
Cognitive behavioural therapy was more often recommended for older patients with more medical and psychiatric conditions, but lower alcohol use. However, the researchers emphasized that the optimal treatment rules were complex and involved intricate interactions between multiple factors that would be difficult for humans to detect or remember.
If implemented in clinical practice, this type of system could potentially help therapists and patients make more informed decisions about treatment approaches. Rather than relying solely on therapist preferences, patient requests, or simple availability, treatment decisions could be informed by data driven predictions about what's most likely to work for each individual.
The researchers found that their algorithm would have assigned more patients to solution focused therapy and fewer to psychodynamic therapy compared to what actually happened in practice. However, every therapy type was still recommended as optimal for some patients, supporting the value of having multiple treatment options available.
Despite these limitations, the study provides compelling evidence that there's room for improvement in how we match patients to therapies. The researchers suggest that even if the real world benefits are smaller than their models predict, population level improvements in mental health treatment could still be substantial.
The most practical application might focus on patients for whom the algorithm makes strong, consistent recommendations across different outcome measures. Rather than trying to optimize therapy assignment for everyone, clinicians could use algorithmic assistance primarily for cases where the predicted benefits are largest and most reliable.
The Human Element Remains
This research doesn't suggest that artificial intelligence should replace clinical judgment or that therapy assignment should be entirely automated. Instead, it points toward a future where data driven insights could complement human expertise to help patients get the most effective treatment for their specific circumstances.
The study also reinforces that having multiple therapy options available is valuable, since different approaches were optimal for different people. This argues against oversimplifying mental health treatment or assuming that one size fits all approaches will work for everyone.
As mental health care continues to evolve, tools like these could help ensure that the right person gets the right treatment at the right time, potentially improving outcomes for the millions of people seeking help for mental health challenges. While we're not there yet, this research provides a compelling glimpse of how technology might enhance the very human work of psychological healing.
Malkki, V. K., Saarni, S. E., Lutz, W., & Rosenstöm, T. H. (2025). Targeted learning for optimal patient assignment to psychotherapy. Psychotherapy Research, 1-15.
The Art of Reading the Room: What Makes Therapy Sessions Actually Work
For nearly four decades, researcher Jeanne Watson has been investigating a deceptively simple question: What exactly happens in those crucial moments during therapy that leads to real change? Her extensive research reveals that successful therapy isn't just about having the right techniques or theories, but about therapists learning to read subtle cues and respond to what's happening in the room moment by moment.
The Dance of Connection
Watson's research shows that both therapists and clients are constantly monitoring the "feel" of their sessions, much like dancers staying attuned to their partner's movements. When things are going well, both participants report a sense of flow and momentum. Clients feel energized and curious, wanting to explore deeper. There's a natural rhythm to the conversation, and both people feel like they're working as a team.
But when sessions hit rough patches, that flow gets interrupted. Clients might feel frustrated or stuck, while therapists sense that their suggestions aren't landing well. The conversation feels forced or jerky, and both parties become painfully aware that something isn't clicking.
The Power of Self Disclosure
One surprising finding is how important client openness is to building strong therapeutic relationships. When clients feel safe enough to share their inner thoughts, feelings, and experiences, it creates a profound sense of connection that both therapist and client can feel. This isn't just about being honest, it's about feeling genuinely understood and accepted.
However, shame can be a major barrier to this openness. Watson's research found that clients who felt ashamed or worried about being judged had much more difficulty forming strong therapeutic relationships. These clients often doubted whether their therapists truly liked or valued them, which made it harder for them to open up and engage fully in the therapeutic process.
When Therapists Push Too Hard
Interestingly, the research reveals that therapists can sometimes become too controlling or directive in their approach, which often backfires. When therapists become overly focused on their own agenda or try to force clients through specific exercises, clients may become resistant, though they often don't express this directly.
Different types of therapy seem to trigger resistance in different ways. In cognitive behavioural therapy, clients were more likely to resist when therapists became too teaching-focused or asked too many directive questions. In emotion-focused therapy, clients pushed back when therapists tried to be overly structured or made interpretations that didn't fit.
The Language of Change
Watson's work also uncovered fascinating patterns in how clients express themselves during productive sessions. When clients use vivid, concrete, and specific language to describe their experiences, they're more likely to access their emotions and have breakthrough moments. This type of detailed storytelling appears to help clients reconnect with their emotions and gain new insights.
The research also found that emotional processing is crucial across different types of therapy, although it manifests differently depending on the approach. In emotion-focused therapy, clients tend to explore their inner experiences more deeply, while in cognitive behavioural therapy, they focus more on external events and logical analysis. Both can be effective, but they engage different aspects of the human experience.
Reading the Signals
Perhaps most importantly, Watson's research emphasizes that effective therapists learn to read countless subtle signals during sessions. They pay attention to changes in vocal quality, shifts in energy, moments of hesitation, and signs that clients are feeling confused or reluctant.
When therapists notice these signals, the most effective response is often to step back from their planned agenda and follow the client's lead more closely. This might mean acknowledging that an exercise isn't working, exploring what the client is actually experiencing in the moment, or simply being more accepting of whatever the client is sharing.
What Makes Some Clients Thrive
The research also revealed clear differences between clients who did well in therapy versus those who struggled. Successful clients were typically able to identify specific problems, actively engage in exploring their experiences, and translate insights into real world changes. They had access to self compassion and could regulate their emotions effectively.
In contrast, clients who had more difficulty often struggled to identify and label their feelings, felt overwhelmed by emotions, and experienced deep shame that made it hard to focus on themselves. They were often more passive in sessions and lacked supportive relationships in their daily lives.
The Therapist's Inner Experience
Watson's work didn't just focus on clients; she also examined what goes on inside therapists' minds during sessions. Effective therapists are constantly balancing multiple streams of information: tracking their client's emotional state, staying aware of their therapeutic framework, and integrating their growing knowledge about the individual client.
When sessions aren't going well, therapists often experience their own difficult emotions, including frustration, confusion, and self doubt. The most effective therapists learn to use these feelings as information rather than getting stuck in them. Instead of becoming defensive or blaming clients, they step back and try to understand what's not working and how to adjust their approach.
This extensive research suggests that therapy effectiveness depends heavily on moment to moment responsiveness rather than rigid adherence to techniques. The most skilled therapists develop a sensitivity to the subtle rhythms of therapeutic conversation and learn to adjust their approach based on what's actually happening rather than what they planned to happen.
The research also highlights the importance of creating safety for clients to be genuinely open about their experiences, including their reactions to therapy itself. When clients feel they can express doubts, confusion, or disagreement without damaging the relationship, it often leads to stronger therapeutic connections and better outcomes.
Watson's decades of research illuminate the incredible complexity of therapeutic change while also providing practical guidance for making therapy more effective. Her work suggests that the future of therapy lies not in developing ever more sophisticated techniques, but in helping therapists become more attuned to the subtle interpersonal processes that either facilitate or hinder healing.
Understanding these moment to moment dynamics can help both therapists and clients recognize when therapy is on track and when adjustments are needed. It reminds us that effective therapy is fundamentally a human endeavor, requiring sensitivity, flexibility, and genuine attunement between two people working together toward healing and growth.
This research reinforces that while different therapeutic approaches may use different techniques, the underlying human processes of connection, understanding, and responsive attunement remain central to helping people change and heal.
Watson, J. C. (2025). Psychotherapy process research: Identifying productive in-session processes to enhance treatment outcomes and therapist responsiveness. Psychotherapy Research, 35(1), 4-16.
