The results of the proposed experiments will directly inform clinicians who treat patients with osteoarthritis of the knee about how to maximize the benefits of acupuncture treatments. And because the experiments specifically asks the question of what is the relation between a patient's expectation of how a treatment will relieve their pain and the outcome of the treatment, the results will potentially inform care givers about all treatments for osteoarthritis and other chronic pain disorders. We hypothesize that acupuncture treatment will produce clinically significant analgesia as indicated by lowered sensory ratings of noxious stimuli and endogenous knee pain.

Inclusion Criteria:

• Volunteers 40-70 years of age.
• Meet the Classification Criteria of the American College of Rheumatology for osteoarthritis of the right and/or left knee for at least the past 3 months, as determined by the referring physician.
• Radiographic evidence of Grade 2 or 3 knee OA using the Kellgren-Lawrence Scale.
• Ability to read and understand English; English can be a second language provided that the patients feel they understand all the questions used in the assessment measures.

Exclusion Criteria:

• Any interventional procedure for knee pain, including corticosteroid injections (within 2 months) to the knee.
• Prior acupuncture treatment for any condition. Because we are using a placebo needle as a control, acupuncture-naive patients are necessary to maximize the benefits of blinding and to control expectancy.
• The intent to undergo surgery during the time of involvement in the study.
• Presence of any illness or medication use that is judged to interfere with the trial. For example: skin irritations around the knee such as psoriasis; bleeding disorders or anticoagulant use that would be contraindications for acupuncture; diabetes due to the increased possibility of sensitivity to heat pain; and use of opiate medications and other substances of abuse that may influence the patient's experience of analgesia. (Due to the potential risk that prescription or non-prescription medication use can confound our results, we may perform a urine toxicology screen to verify patient's medication status during Session 1.)
• Knee pain due to other causes, such as inflammation or malignancy, other pain disorders that may refer pain to the leg, OA of ipsilateral hip, diagnosis of rheumatoid arthritis (RA).
• Non-ambulatory status.
• History of cardiac, respiratory, or nervous system disease that, in the investigator's judgment, precludes participation in the study because of a heightened potential for adverse outcome. For example: asthma or claustrophobia.
• Presence of any contraindications to fMRI scanning. For example: cardiac pacemaker, metal implants, fear of closed spaces, pregnancy.
• Instability of pain rating within Session 1 or Session 2 of Experiment One or Visit 1 of Experiment Two.