What is muscle testing?
Muscle testing is a way of reading the body's response as a yes or a no. In the most common form, a person holds an arm out while the practitioner presses lightly on the wrist and asks a question or has the person make a statement. If the arm holds firm, that is read as yes or true. If it gives way a little, that is read as no or false.
Discover Healing describes muscle testing as a way to ask the subconscious questions and receive answers about what the body indicates. Its materials teach several forms: the arm test, a sway test in which a person stands still and notices whether they drift forward or backward, and finger tests a practitioner can do on themselves. Proxy testing, in which one person is tested on behalf of another, is also taught.
In the Emotion Code, muscle testing is how a practitioner decides whether there is a trapped emotion, which emotion it is, and whether it has been released. So the accuracy of muscle testing matters for everything else the method claims.
Where does muscle testing come from?
Muscle testing in this sense comes from applied kinesiology, developed by the chiropractor George Goodheart in the 1960s. Applied kinesiology proposed that changes in muscle strength could reveal problems elsewhere in the body, such as nutritional needs or sensitivities to substances. Many later methods, including the Emotion Code, borrowed the technique.
It is important not to confuse this with manual muscle testing as used by physical therapists and doctors, which grades real muscle strength after an injury or in a neurological exam. That kind of testing measures the muscle itself. Applied kinesiology uses the muscle as a signal about something else, and that is the claim researchers have tested.
What have controlled studies found?
The fair way to test muscle testing is to blind it: neither the tester nor the person being tested knows the right answer, so neither can influence the result, even unconsciously. Several studies have done this. Here is what the main ones found.
Nutrient status (1988)
Kenney and colleagues had three experienced applied kinesiologists assess 11 people for four nutrients and compared the results with laboratory tests. The testers did not agree with one another, their results did not match the lab tests, and in a double-blind phase they responded no differently to nutrients than to placebos. The authors concluded that applied kinesiology was no more useful than random guessing for assessing nutrient status.
Insect venom allergy (2001)
Lüdtke and colleagues asked four kinesiology examiners to test seven people with confirmed wasp venom allergy against 10 venom and 10 placebo bottles, with everyone blinded. Agreement was almost nil (kappa 0.03), and the test picked out the venom only about as often as chance would predict. The authors concluded it was no more useful than random guessing.
Dental materials (2005)
Staehle and colleagues had two dentists trained in applied kinesiology test volunteers for tolerance to dental materials, first openly and then blinded. In the blinded retest, the results matched the open results in 14 of 40 cases, which the authors found consistent with chance.
A hidden toxin (2014)
Schwartz and colleagues ran a double-blind, randomized study with 51 participants who held vials of either a toxic solution or saline while three kinesiologists, and separately a grip-strength dynamometer, tested them. The toxic vial was identified correctly in 53 percent of trial sets, close to the 50 percent expected by chance. The authors concluded that in studies meeting accepted scientific standards, applied kinesiology has not shown it is a useful or reliable diagnostic tool.
Detecting spoken lies (2016 and 2019)
Anne Jensen and colleagues at the University of Oxford tested whether muscle testing could tell true spoken statements from false ones. In two blinded experiments, practitioners were correct about 66 percent of the time, better than chance and better than their own intuitive guesses, which were about 47 percent. A 2019 follow-up again found accuracy above chance. The authors noted that their results do not generalize to other uses of muscle testing. Even at its best here, about one answer in three was wrong.
A 2025 systematic review
Soares and colleagues reviewed seven studies of muscle-testing reliability in applied kinesiology. Reliability ranged from none to strong depending on the muscle and the method. Testing that involved nonmusculoskeletal challenges, such as the substance and question tests applied kinesiology is known for, showed no reliability, and the authors said it is not recommended for clinical use.
What does that mean for the Emotion Code?
Taken together, the research does not support using muscle testing to detect things the tester cannot otherwise know. The Emotion Code uses it for exactly that: to identify a hidden emotion, when it became trapped, and whether it has gone.
There is also a deeper problem. The lie-detection studies could check answers against a known truth. Nobody can check whether a particular trapped emotion is really present, because there is no independent measure of trapped emotions. So muscle testing in the Emotion Code cannot be tested for accuracy in the usual way, and no study has tried. The underlying mechanism the method describes is not established by scientific evidence. Does the Emotion Code work? and questions from skeptics take this further.
Why does muscle testing feel so convincing?
People who experience muscle testing often find it striking: the arm really does seem to weaken on some statements and hold on others. That experience is genuine. The question is what causes it.
Tiny changes in how hard the tester presses, how the person braces, or what either of them expects can change the result without anyone intending it. This is why blinding matters so much. When studies remove the chance for expectations to leak in, the striking effect usually disappears. None of this implies anyone is being dishonest. It is simply how small, unconscious movements work.
How should muscle testing be used, if at all?
- Never use it to make medical decisions, such as whether to take a medication, whether a food is safe for an allergy, or whether a symptom needs a doctor.
- Treat any answer it gives as part of a reflective experience, not as information about your body.
- Be cautious of any practitioner who uses muscle testing to diagnose a condition or to tell you what caused a symptom.
- It is fine to ask a practitioner how they think about accuracy. Laura welcomes the question; ask her through the Emotion Code page.
Common questions
Is muscle testing the same as applied kinesiology?
Muscle testing is the core technique of applied kinesiology, a practice developed by chiropractor George Goodheart in the 1960s. Many methods, including the Emotion Code, use a version of it. It should not be confused with the manual muscle testing physical therapists use to grade real muscle strength, which measures the muscle itself rather than using it as a signal about something else.
Can I muscle test myself?
Discover Healing teaches self-testing methods, such as the sway test and finger tests. Self-testing faces the same problem as testing by a practitioner, only more so: the person asking the question also produces the response, so expectations have an easy way in. If you use it, treat the answers as a prompt for reflection, never as a guide to health decisions.
Does the lie-detection study prove muscle testing works?
No. Jensen and colleagues found muscle testing distinguished spoken lies from truths better than chance, about 66 percent correct, in a specific laboratory setup. The authors said the result does not generalize to other uses. It says nothing about detecting trapped emotions, and it still means about a third of answers were wrong. Other blinded studies of different uses have found chance-level results.
Why do practitioners keep using it if studies are negative?
Practitioners generally find it useful in practice and see the striking arm response as meaningful, and the method's framework treats it as a way of communicating with the subconscious. Experience in a session is persuasive in a way that trial data often is not. The honest position is that muscle testing has not been shown to be a reliable detector, and anyone using it should say so.