Opinion by Judge DAVID R. THOMPSON; Dissent by Judge NOONAN.
DAVID R. THOMPSON, Circuit Judge:Plaintiff-appellant David Morgan (“Morgan”) appeals the district court’s judgment affirming the Commissioner’s denial of Disability Insurance Benefits (“DIB”) under Title XVIII of the Social Security Act, 42 U.S.C. § 423 and Supplemental Security Income (“SSI”) under Title XVI of the Social Security Act, 42 U.S.C. § 1381 et seq.
Morgan contends the Administrative Law Judge (“ALJ”) erred in rejecting Morgan’s testimony as to his mental condition, and in rejecting the opinions of his treating psychiatrist and examining psychologist in favor of the opinion of a nonexamining medical advis- or.
We conclude that the ALJ gave clear and convincing reasons, supported by the record, for rejecting Morgan’s testimony. We also conclude that the ALJ gave specific, legitimate reasons for rejecting the opinions of Morgan’s treating psychiatrist and examining psychologist, and that those reasons are based on substantial evidence in the record. Accordingly, we affirm the denial of benefits.
BACKGROUND
Born in 1950, David Morgan is a high school graduate who also earned an Associate of Arts degree in math and history. Morgan worked for eleven years as a grocery store retail clerk until he was fired in 1988 because of a fight with a supervisor. He has not been employed since 1988. In February 1994, after threatening suicide, Morgan entered the Lane Psychiatric Hospital. During his hospitalization, Morgan reported a recent history of drinking two to three beers per day, and admitted to cannabis-marihuana use. Upon discharge, Morgan’s medical examiners diagnosed him as suffering from a major depressive disorder, alcohol abuse,' cannabis abuse, a history of amphetamine and opioid abuse, and a schizoid and/or dependent personality.
Between March 1994 and March 1995, Dr. Michael Reaves, M.D., evaluated Morgan’s mental health at least six times. He observed Morgan to be “relatively articulate, intelligent [and] ... to have a rather quick wit about him.” Dr. Reaves repeatedly diagnosed Morgan with recurrent major depression, alcohol abuse, cannabis abuse, and provisional schizoid personality disorder. Dr. Reaves rated Morgan’s Global Assessment of *598Functioning (GAF) at 45/55.1 Dr. Reaves prescribed the continued use of antidepres-sive medication.
Between March 1994 and March 1995, Dr. Sally Grosscup, Ph.D., also performed a series of psychological evaluations of Morgan. Dr. Grosscup diagnosed Morgan with “Major Depressive Disorder, single episodes; severe, without psychotic features” and concluded that he had “a moderate to high suicide risk.” In addition, she found repeatedly that Morgan suffered from alcohol dependence, cannabis abuse, amphetamine abuse, opioid abuse, nicotine dependence, inhalant abuse, and a probable borderline personality disorder. Dr. Grosscup assigned Morgan a GAF score of 49, with a high of 61 during the previous year. In one evaluation, Dr. Grosscup opined that “[Morganj’s prognosis for change is extremely poor, given his probable character disorder. His condition has been long term and so far he only moderately responds to medication. He is probably functioning at the highest level of his current capacity.”
Morgan applied for DIB and SSI on March 18,1994, around the time Dr. Reaves and Dr. Grosscup began evaluating him. After his applications were denied initially and upon reconsideration, Morgan requested a disability hearing before an ALJ.
At the April 12, 1995 disability hearing, Dr. William MeConochie, Ph.D., testified as a medical expert. Dr. MeConochie stated that the medical evidence demonstrated a history of major depression “characterized by less than four years of the several symptoms listed, but including a sleep disturbance, difficulty concentrating, and thoughts of suicide.” Dr. MeConochie recognized that other medical examiners made a provisional diagnosis of personality disorder, but stated that he believed the medical evidence did not support this diagnosis. He stated that he agreed with the various professionals who considered Morgan an alcoholic, and that he considered Morgan’s alcoholism and cannabis abuse the most important diagnoses. Dr. MeCono-chie noted that, while Morgan’s depression probably played a part in the primary diagnosis, it was more likely a side effect of the alcoholism and cannabis abuse. When asked about the “B” criteria of listing 12.00 of the Social Security Regulations,2 Dr. MeConochie testified that Morgan did not meet the listings, having only slight difficulties in maintaining social functioning and never or seldom having deficiencies of concentration, persistence, or pace. Dr. MeConochie also testified that Morgan had no significant limitations on a Mental Residual Functional Capacity assessment, contrary to the opinions of Dr. Reaves and Dr. Grosscup.
The ALJ issued a decision on August 24, 1995, denying Morgan’s claim. In reaching his decision, the ALJ concluded that Morgan’s allegations, both in his testimony and in his statements to the medical examiners, were not highly credible. The ALJ determined that clear conflict existed between Morgan’s statements regarding the severity of his depression as he represented those symptoms and Morgan’s behavior as observed under objective testing. This conflict between subjective testimony and objective medical evidence led the ALJ to conclude that Morgan’s subjective complaints reflected a lack of credibility and a degree of exagger*599ation. The ALJ discounted, therefore, any medical findings based on Morgan’s subjective complaints, including the medical examiners’ opinions regarding his ability to work.
Based on these determinations and other evidence in the record, the ALJ concluded that Morgan had severe impairments, but that the impairments did not meet or equal the requirements for disability, either individually or in combination. The ALJ concluded that Morgan retained the residual functional capacity to perform his past relevant work as a retail clerk, and therefore was not disabled within the meaning of the Social Security Act.
After the Appeals Council denied review of Morgan’s claim, thereby making the ALJ’s determination a final decision of the Commissioner, Morgan filed a complaint asking the United States District Court for the District of Oregon to review and set aside the Commissioner’s decision. On November 7, 1997, United States District Judge Michael R. Hogan adopted the findings and recommendation of Magistrate Judge John P. Cooney, affirming the Commissioner’s denial of Morgan’s applications for benefits.
Morgan timely appeals the district court’s decision. We have jurisdiction under 28 U.S.C. § 1291, and we affirm.
STANDARD OF REVIEW
We review de novo the district court’s order affirming the Commissioner’s denial of benefits. See Travers v. Shalala, 20 F.3d 993, 995-96 (9th Cir.1994). The Commissioner’s decision to deny benefits will be overturned “only if it is not supported by substantial evidence or is based on legal error.” Magallanes v. Bowen, 881 F.2d 747, 750 (9th Cir.1989) (quoting Green v. Heckler, 803 F.2d 528 (9th Cir.1986)). Substantial evidence means such relevant evidence as a reasonable mind might accept as adequate to support a conclusion. Id. Where the evidence is susceptible to more than one rational interpretation, it is the ALJ’s conclusion that must be upheld. Andrews v. Shalala, 53 F.3d 1035, 1041 (9th Cir.1995).
DISCUSSION
Morgan contends the ALJ failed to provide clear and convincing reasons for rejecting Morgan’s testimony of subjective symptoms. Morgan also argues the ALJ rejected the opinions of his examining psychologist and treating psychiatrist without providing specific and legitimate reasons. We disagree.
For the ALJ to reject Morgan’s subjective complaints, he must provide “specific, cogent reasons for the disbelief.” Lester v. Chater, 81 F.3d 821, 834 (9th Cir.1995) (quoting Rashad v. Sullivan, 903 F.2d 1229, 1231 (9th Cir.1990)). Without affirmative evidence showing that the claimant is malingering, the Commissioner’s reasons for rejecting the claimant’s testimony must be clear and convincing. See Lester, 81 F.3d at 834; Swenson v. Sullivan, 876 F.2d 683, 687 (9th Cir.1989). If an ALJ finds that a claimant’s testimony relating to the intensity of his pain and other limitations is unreliable, the ALJ must make a credibility determination citing the reasons why the testimony is unpersuasive. See Bunnell v. Sullivan, 947 F.2d 341 (9th Cir.1991). The ALJ must specifically identify what testimony is credible and what testimony undermines the claimant’s complaints. See Lester, 81 F.3d at 834; Dodrill v. Shalala, 12 F.3d 915, 918 (9th Cir.1993); Varney v. Secretary of Health and Human Services, 846 F.2d 581, 584 (9th Cir.1988). In this regard, questions of credibility and resolutions of conflicts in the testimony are functions solely of the Secretary. See Yuckert v. Bowen, 841 F.2d 303, 307 (9th Cir.1988); Sample v. Schweiker, 694 F.2d 639, 642 (9th Cir.1982).
The ALJ provided clear and convincing reasons for rejecting Morgan’s testimony. The ALJ pointed to specific evidence in the record — including reports by Dr. Reaves, the treating psychiatrist, and Dr. Grosscup, the examining psychologist — in identifying what testimony was not credible and what evidence undermined Morgan’s complaints. F or example, the ALJ noted that, contrary to Morgan’s claims of lack of improvement, Dr. Reaves reported that Morgan’s mental symptoms improved with the use of medication. The ALJ also commented upon contradictions between Morgan’s reports to Dr. *600Grosscup and his reports to Dr. Reaves regarding his activities. The ALJ also identified inconsistency among the Lane County Mental Health staffs observations and Morgan’s denial of improvement in his condition and claims of contemplated suicide.
Morgan argues that the ALJ incorrectly determined that there was no evidence of his concentration problems in the psychological reports. Morgan points to several occasions when he complained of concentration problems to his examining physicians. Rather than determining that Morgan did not complain of concentration problems or that they did not exist in the record, the ALJ found that the examining psychologist’s and treating psychiatrist’s opinions were not supported by Morgan’s complaints. The ALJ recognized that although Morgan “indicated that he had significant concentration problems and that he tends to ‘drift’ while doing tasks on a sustained basis, even yard work, an endeavor he previously acknowledged he could do well ..., his mental status evaluations did not reveal such cognitive defects. On the contrary, they revealed superior intellectual ability.”
Contrary to Morgan’s claims of disability, the ALJ determined that Morgan’s ability to fix meals, do laundry, work in the yard, and occasionally care for his friend’s child served as evidence of Morgan’s ability to work. If a claimant is able to spend a substantial part of his day engaged in pursuits involving the performance of physical functions that are transferable to a work setting, a specific finding as to this fact may be sufficient to discredit a claimant’s allegations. See Fair v. Bowen, 885 F.2d 597, 603 (9th Cir.1989). Rather than misconstruing the examining physician’s diagnosis of a “lack of motivation,” as Morgan contends, the ALJ, in noting Morgan’s daily activities, pointed to the contradictions between Morgan’s reported activities and his asserted limitations as an issue of credibility.
Citing the conflict between Morgan’s testimony of subjective complaints and the objective medical evidence in the record, and noting the ALJ’s personal observations, the ALJ provided specific and substantial reasons that undermined Morgan’s credibility. “The inclusion of the ALJ’s personal observations does not render the decision improper.” Sellard v. Shalala, 37 F.3d 1506 (9th Cir.1994), citing Nyman v. Heckler, 779 F.2d 528, 531 (9th Cir.1985). “Where, as here, the ALJ has made specific findings justifying a decision to disbelieve an allegation ... and those findings are supported by substantial evidence in the record, our role is not to second-guess that decision.” Fair, 885 F.2d at 603.
Morgan argues that the ALJ erred in rejecting the opinions of his treating psychiatrist and examining psychologist in favor of the nonexamining medical advisor. We reject this argument, however, because the ALJ gave legitimate and specific reasons, supported by the record, for rejecting the opinions of the treating psychiatrist and examining psychologist.
The opinion of a treating physician is given deference because “he is employed to cure and has a greater opportunity to know and observe the patient as an individual.” Sprague v. Bowen, 812 F.2d 1226, 1230 (9th Cir.1987) (citations omitted). However, the opinion of the treating physician is not necessarily conclusive as to either the physical condition or the ultimate issue of disability. See Magallanes v. Bowen, 881 F.2d 747, 751 (9th Cir.1989); Rodriguez v. Bowen, 876 F.2d 759, 761-62 & n. 7 (9th Cir.1989). When a nontreating physician’s opinion contradicts that of the treating physician-but is not based on independent clinical findings, or rests on clinical findings also considered by the treating physician-the opinion of the treating physician may be rejected only if the ALJ gives “specific, legitimate reasons for doing so that are based on substantial evidence in the record.” Andrews v. Shalala, 53 F.3d 1035, 1041 (9th Cir.1995); Magallanes, 881 F.2d at 751, 755 (quoting Sprague, 812 F.2d at 1230). Opinions of a nonexamin-ing, testifying medical advisor may serve as substantial evidence when they are supported by other evidence in the record and are consistent with it. Andrews, 53 F.3d at 1041. “The ALJ can meet this burden by setting out a detailed and thorough summary of the facts and conflicting clinical evidence, *601stating his interpretation thereof, and making findings.” Magallanes, 881 F.2d at 750, (quoting Cotton v. Bowen, 799 F.2d 1403, 1408 (9th Cir.1986)). Where medical reports are inconclusive, “questions of credibility and resolution of conflicts in the testimony are functions solely of the Secretary.” Sample, 694 F.2d at 642 (quoting Waters v. Gardner, 452 F.2d 855, 858 n. 7 (9th Cir.1971)).
Substantial evidence supports the ALJ’s determination that Dr. Grosscup’s reports do not show how Morgan’s symptoms translate into specific functional deficits which preclude work activity. Although Dr. Grosscup identified characteristics that might limit Morgan’s ability to work on a sustained basis: affective instability, intense anger, daily suicidal thoughts, and chronic feelings of emptiness, Dr. Grosscup did not explain how these characteristics precluded work activity in Morgan’s case. The ALJ listed specific examples of how the level of impairment indicated by Dr. Grosscup was unreasonable given the description of Morgan’s symptoms in her reports and other evidence in the record. For example, Dr. Grosscup, at various times, assessed Morgan as engaging and cooperative, with a sense of humor and a good ability to express himself. She indicated that Morgan was adequately groomed and punctual for his appointments. The ALJ relied on these specific factors in rejecting Dr. Grosseup’s opinion that Morgan “lacked an ability to function socially in order to conduct personal business and otherwise carry out daily activities.”
Morgan contends that Dr. Grosscup’s diagnoses of the effectiveness of Morgan’s medication and of Morgan’s suicidal impulses are consistent with the views of the other medical examiners. But the ALJ noted that during July 1994, Dr. Reaves recognized “significant improvement” from the medication, while Dr. Grosscup found only “moderate change.” Dr. Reaves determined that Morgan was “not currently actively suicidal,” while Dr. Grosscup noted “current suicidal thoughts.” At best, the doctors’ almost contemporaneous diagnoses are inconclusive. Where medical reports are inconclusive, “questions of credibility and resolution of conflicts in the testimony are functions solely of the [Commissioner].” Sample, 694 F.2d at 642 (quoting Waters, 452 F.2d at 858 n. 7). Consequently, when evidence is susceptible to more than one rational interpretation, the ALJ’s conclusion must be upheld. See Andrews, 53 F.3d at 1041; Magallanes, 881 F.2d at 750.
Next, Morgan argues the ALJ substituted his own judgment for that of medical sources in rejecting Dr. Grosscup’s diagnosis of alcohol dependence. Morgan notes that all three medical experts, including Dr. McConochie, the nontreating, nonexamining medical expert who testified at the hearing, diagnosed Morgan with some form of “Alcoholism.”3
Morgan correctly asserts that an ALJ must accompany rejection of an expert’s un-controverted opinion with “clear and convincing” reasons for doing so. Andrews, 53 F.3d at 1041; Magallanes, 881 F.2d at 751. But rather than rejecting Dr. Grosscup’s diagnosis of alcohol dependence, the ALJ rejected Morgan’s attempt to establish disability under the Social Security Act based on alcoholism. The ALJ noted that Morgan held his previous job while consuming alcohol at the level he contends established his alcohol dependency.
In challenging the ALJ’s rejection of Dr. Reaves’s opinions, Morgan argues the ALJ wrongly relied upon the circumstance that Dr. Reaves first examined Morgan in 1994, yet he diagnosed Morgan as disabled from 1989, four years earlier. Morgan argues there was nothing improper about Dr. Reaves’s retroactive diagnosis.
We agree with Morgan that the circumstance of a retroactive diagnosis, standing alone, may not be sufficient to discount the opinion of a treating physician. Here, however, there was more. The ALJ noted, contrary to Dr. Reaves’s conclusion that Morgan suffered from “marked limitations,” that the evidence established that Morgan adequately copes with the social aspects of daily living, *602continues to maintain some friendships, and manifests above-average intelligence and other cognitive abilities. In addition, although Dr. Reaves determined that Morgan could not work without interruptions from psychologically based symptoms, perform at a consistent pace without an unreasonable number and length of rest periods, and work in coordination or proximity to others without being distracted by them, the ALJ recognized that no substantial evidence existed demonstrating that Morgan’s mental impairments prevented him from working. In fact, the ALJ noted that Morgan demonstrated the ability to work independently and be self-motivated by maintaining a one-acre garden and assisting with the restoration of an old house.
Inconsistency between Dr. Grosscup’s and Dr. Reaves’s conclusions provided the ALJ additional justification for rejecting Dr. Reaves’s opinion. The ALJ found that Dr. Reaves’s opinions regarding Morgan’s cognitive abilities contradicted those offered by Dr. Grosscup in her mental status evaluation. Further, Dr. Reaves’s conclusion regarding Morgan’s suicide risk was inconsistent with that of Dr. Grosscup. On the other hand, the ALJ noted that Dr. Reaves, like Dr. Grossc-up, allowed Morgan’s attitude and lack of interest in work to influence his assessment that Morgan could not work.
The ALJ also rejected the opinions of Dr. Grosscup and Dr. Reaves in part because those opinions were premised on Morgan’s subjective complaints, which the ALJ discounted for the reasons previously stated. A physician’s opinion of disability “premised to a large extent upon the claimant’s own accounts of his symptoms and limitations” may be disregarded where those complaints have been “properly discounted.” Fair, 885 F.2d at 605, citing Brawner v. Secretary of Health & Human Servs., 839 F.2d 432, 433-34 (9th Cir.1988).
In rejecting the opinions of Dr. Grosscup and Dr. Reaves, the ALJ also relied on the opinion of Dr. McConochie, the nontreating, nonexamining, testifying medical expert. Dr. McConochie testified that the medical evidence did not support a diagnosis of personality disorder. Dr. McConochie concluded that Morgan did not meet the requirements for disability, because Morgan, had only slight difficulties in maintaining social functioning and never or seldom had deficiencies of concentration, persistence or pace. In contrast with the opinions of Dr. Grosscup and Dr. Reaves, Dr. McConochie testified that Morgan had no significant limitations on the MFRC assessment.
The opinion of a nonexamining medical advisor cannot by itself constitute substantial evidence that justifies the rejection of the opinion of an examining or treating physician. See Lester, 81 F.3d 821; Pitzer v. Sullivan, 908 F.2d 502, 506 n. 4 (9th Cir.1990); Gallant v. Heckler, 753 F.2d 1450, 1456 (9th Cir.1984). In Gallant, we determined that “the report of [a] nontreating, nonexamining physician, combined with the ALJ’s own observation of [the] claimant’s demeanor at the hearing,” did not constitute substantial evidence and, therefore, did not support the Commissioner’s rejection of the examining physician’s opinion that the claimant was disabled. Gallant, 753 F.2d at 1456. In Pitzer, we held that the nonexamining physician’s opinion “with nothing more” did not constitute substantial evidence. Pitzer, 908 F.2d at 502.
But we have consistently upheld the Commissioner’s rejection of the opinion of a treating or examining physician, based in part on the testimony of a nontreating, nonexamining medical advisor. See, e.g., Magallanes, 881 F.2d at 751-55; Andrews, 53 F.3d at 1043; Roberts v. Shalala, 66 F.3d 179 (9th Cir.1995). In Magallanes, evidence that supported the ALJ’s determination included, among other things, testimony from the claimant that conflicted with her treating physician’s opinion. Magallanes, 881 F.2d at 751-51. In Andrews, the fact that the results of psychological testing conducted by the examining psychologist were suspect, as a product of the claimant’s drug use, contributed to the ALJ’s findings. Andrews, 53 F.3d at 1035.
In the present case, the ALJ pointed to specific evidence in addition to his personal observations, and the opinion of the nontreat-ing, nonexamining medical advisor, Dr. McConochie, to support his rejection of the *603opinions of Dr. Reaves and Dr. Grosscup. As in Magallanes, the ALJ cited to testimony from Morgan that conflicted with the Reaves and Grosscup opinions. Similar to Andrews, the ALJ in the present case discounted the results of the psychological testing conducted by the examining psychologist based on Morgan’s alcohol use and his “not entirely credible” testimony.
Moreover, internal inconsistencies within Dr. Reaves’s and Dr. Grosscup’s reports, and the inconsistencies between their reports, also constitute relevant evidence. The ALJ is responsible for resolving conflicts in medical testimony, and resolving ambiguity. See Andrews, 53 F.3d at 1041; Magallanes, 881 F.2d at 751, 755. Determining whether inconsistencies are material (or are in fact inconsistencies at all) and whether certain factors are relevant to discount the opinions of Dr. Reaves and Dr. Grosscup falls within this responsibility.
CONCLUSION
We affirm the judgment of the district court, which upheld the final decision of the Commissioner denying Morgan’s application for Disability Insurance Benefits under Title XVIII of the Social Security Act, 42 U.S.C. § 423 and Supplemental Security Income under Title XVI of the Social Security Act, 42 U.S.C. § 1381 et seq.
AFFIRMED.
. Clinicians use a GAF to rate the psychological, social, and occupational functioning of a patient. The scale does not evaluate impairments caused by psychological or environmental factors. A GAF between 41 and 50 indicates serious symptoms (e.g. suicidal ideation, severe obsessional rituals, frequent shoplifting) or any serious impairment in social, occupational, or school functioning (e.g., no friends, unable to keep a job).
. The "B” criteria lists those functional limitations that claimants must demonstrate to meet the mental disorder listings. To meet Listing 12.04, Affective Disorders, a claimant's disturbance in mood must result at least two of the following:
(1) marked restriction of activities of daily living;
(2) marked difficulties in maintaining social functioning;
(3) deficiencies of concentration, persistence or pace resulting in frequent failure to complete tasks in a timely manner (in work settings or elsewhere); or
(4) repeated episodes of deterioration or de-compensation in work or work-like settings which cause the individual to withdraw from that situation or to experience exacerbation of signs and symptoms (which may include deterioration of adaptive behaviors).
See 20 C.F.R. Part 404, Subpart P, Appendix 1, SS 12.04B(l)-(4).
. To be exact, Dr. Grosscup diagnosed Morgan as suffering from Alcohol Dependence, Dr. Reaves diagnosed Morgan as suffering from Alcohol Abuse, and Dr. McConochie diagnosed Morgan as suffering from Alcoholism.