Jerry C. VALENTINE, Plaintiff-Appellant, v. COMMISSIONER SOCIAL SECURITY ADMINISTRATION, Defendant-Appellee, 574 F.3d 685

O’SCANNLAIN, Circuit Judge: We must determine the circumstances under which the Social Security Administration can find that a claimant for disability insurance benefits is not disabled despite a contrary finding by the Department of Veterans Affairs. I Jerry Valentine filed an application with the Social Security Administration (“SSA”) for Social Security disability insurance benefits in March 2005. He alleged disability beginning the previous year. A former member of the Navy, Valentine suffered a combat head injury in Vietnam in 1969.

Since then, he has held several jobs, most recently as a “parts man” for a diesel engine distributor called Cummins Northwest Incorporated. In March 2004, he retired from Cummins. Valentine complains of post-traumatic stress disorder (“PTSD”), a combination of depression and sleep deprivation (sometimes referred to in the record as cognitive disorder secondary sleep deprivation), and degenerative joint disease in his right shoulder and left knee.

The PTSD resulted from Valentine’s head injury in Vietnam, but it appears to have worsened significantly in the wake of the death of his brother from a head injury in the summer of 2000. Valentine received treatment for the PTSD and for sleep disturbance and persistent nightmares at the Veterans Administration Medical Center in Portland, Oregon, from September of 2000 through the date of the hearing before the Administrative Law Judge (“ALJ”). While undergoing sleep therapy in 2001, Valentine began treatment with Dr. Lynn M. Van Male.

At the time, Valentine reported that he got a good night’s sleep about three days out of each week. Valentine’s performance at work was erratic; he received several “marginal” performance ratings on his annual job review in February 2002. Dr. Van Male referred Valentine to Dr. Daniel Storzbach for a neuropsychological assessment. The results “suggested average baseline cognitive ability”; performance on some of the specific tests indicated “normal limits or better.” However, other tests indicated impaired performance with respect to attention, working memory, and complex pyschomotor function. Dr.

Storzbach seemed to attribute these difficulties to the exacerbation of Valentine’s PTSD symptoms following his brother’s death. He recommended several ways to cope and noted that Valentine would probably have less difficulty with “highly routinized, over-learned tasks with low cognitive demand.” Meanwhile, Valentine, with Dr. Van Male’s help, tried to increase his disability rating, which then stood at 30 percent, from the Department of Veterans Affairs (“VA”). In a letter to the VA written in May 2002, Dr.

Van Male stated that Valentine had tried to hold down his job “at significant cost to himself,” but that she worried about his ability to maintain employment “given his current rate of functional decline.” At the time, Valentine was reporting increased mental stability at work and at home (for example, he was able to garden with his wife), despite his sleeping problems. The VA raised his disability rating to 70 percent. Though difficulties persisted during 2003, Valentine managed them reasonably well. He continued to work and paid off credit card and truck loan debts.

His performance review in January 2003 was positive, his ratings being in the “acceptable” to “commendable” range. Whether because his fatigue and associated ailments became too much for him or because he became eligible to receive his employee pension, Valentine planned to retire in March 2004. Things improved as Valentine’s planned retirement date approached. He received positive reviews in February from his supervisor, Lane Anthony. Anthony praised Valentine’s attitude as “outstanding,” called him “a wonderful asset to the company’s front counter,” and noted improvement in his work product.

Valentine assured company executives that he was ready and willing to retire, despite their offers of a shortened or split shift, because he believed it was in his best interests. After his retirement, Valentine’s condition continued to improve as he exercised and took up projects to keep busy. He even stopped regular visits with Dr. Van Male in November 2004. Valentine requested another increase in his disability rating from the VA. Dr. Leslie Carter interviewed him in October 2004 and found “well-documented” Valentine’s assertion that his nightmares and sleep deprivation were “extremely disabling.” Dr.

Carter, however, was under the impression that Valentine had quit working at Cummins because he was about to be fired. Initially the VA did not act, despite Dr. Carter’s report. But after Dr. Van Male sent farther letters in 2005 and 2006, the VA ultimately raised Valentine’s disability rating to 100 percent. In addition to his PTSD, Valentine sustained two physical injuries in 2005. He tore some cartilage in his shoulder and damaged his left knee. He underwent surgery for both injuries.

In September 2005, two months before surgery on his knee, Valentine took a physical examination, which did not suggest any significant physical impairment. Several psychologists, including a Dr. Peter LeBray, reviewed Valentine’s medical record on behalf of the SSA. The ALJ considered this evidence, along with the rest of Valentine’s file, at a hearing in March of 2006. Ultimately, the ALJ decided that Valentine was not disabled and denied him benefits. The Appeals Council declined review, making the ALJ’s decision the final decision of the Commissioner.

Valentine filed a civil action in the district court to obtain judicial review of the agency’s decision. The district court affirmed the denial of benefits, and Valentine now appeals. II To establish eligibility for Social Security disability benefits, a claimant has the burden to prove he is disabled. See Roberts v. Shalala, 66 F.3d 179, 182 (9th Cir. 1995). Like most Social Security cases, this case involves the agency’s five-step procedure for determining disability. See 20 C.F.R. § 404.1520(a)(4)(i)-(v); see also Bustamante v. Massanari, 262 F.3d 949, 953-54 (9th Cir.2001).

The ALJ’s decision turned on her conclusion at step five. While the claimant has the burden of proof at steps one through four, “the burden of proof shifts to the [Commissioner]” at step five “to show that the claimant can do other kinds of work.” Embrey v. Bowen, 849 F.2d 418, 422 (9th Cir.1988). To direct this inquiry, the Commissioner, through the ALJ, must determine the claimant’s “residual functional capacity,” a

summary of what the claimant is capable of doing (for example, how much weight he can lift). The ALJ may, and did here, pose to a vocational expert a hypothetical incorporating the residual functional capacity determination (“RFC”); the expert then opines on what kind of work someone with the limitations of the claimant could hypothetically do. See Roberts, 66 F.3d at 184. The ALJ must then determine whether, given the claimant’s RFC, age, education, and work experience, he actually can find some work in the national economy. See Tackett v.

Apfel, 180 F.3d 1094, 1100-01 (9th Cir.1999); 20 C.F.R. § 404.1520(a)(4)(v). We “review[] the district court’s order affirming the [Commissioner]^ denial of benefits de novo ... to ensure that the [Commissioner]’s decision was supported by substantial evidence and a correct application of the law.” Roberts, 66 F.3d at 182 (internal citation omitted). This is a highly deferential standard of review: “ ‘Substantial evidence’ means more than a mere scintilla, but less than a preponderance. It means such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.” Desrosiers v.

Sec’y of Health & Human Servs., 846 F.2d 573, 576 (9th Cir.1988) (internal quotation marks and citations omitted). Ill Before reviewing the details of the ALJ’s decision, we must dispose of a threshold issue. Valentine alleges that the ALJ denied him due process “because of the ALJ’s attitude and demeanor.” He claims that several of the ALJ’s pointed questions and expressions of disbelief made the administrative hearing less than full and fair. Valentine does not allege that the ALJ was biased against him. Instead, he merely suggests that the ALJ had prejudged his ease in some way.

We can find no legal authority for the proposition that general preconceptions that do not amount to bias violate the Due Process Clause. Even if we construe Valentine’s complaint to allege actual bias, he has mustered no evidence that comes close to the required showing. ALJs are presumed to be unbiased. Rollins v. Massanari, 261 F.3d 853, 857 (9th Cir.2001). Such presumption “can be rebutted by a showing of conflict of interest or some other specific reason for disqualification. ...

But expressions of impatience, dissatisfaction, annoyance, and even anger, that are within the bounds of what imperfect men and women sometimes display!,] do not establish bias.” Id. at 857-58 (internal quotation marks omitted). That the ALJ questioned some of Valentine’s arguments and the perspective of some of those who treated him plainly does not show bias in this sense. IV We proceed to Valentine’s more specific objections, the first of which concerns the RFC. Valentine contends that the ALJ did not take account of all his limitations in fashioning the RFC.

The hypothetical an ALJ poses to a vocational expert, which derives from the RFC, “must set out all the limitations and restrictions of the particular claimant.” Embrey, 849 F.2d at 422. Thus, an RFC that fails to take into account a claimant’s limitations is defective. Valentine’s objections to the RFC can be divided into two categories. He argues that the ALJ ignored some and improperly rejected other evidence of the extent of his impairments.

A The ALJ concluded that Valentine “has the residual functional capacity to perform a limited range of medium exertion work.” Specifically, the RFC reads: [Valentine] is able to lift up to 50 pounds. He is able to sit at least six hours out of an eight hour workday. He is able to stand at least six hours out of an eight hour workday. He is able to perform postural activities frequently. He has moderate restrictions of his capacity to concentrate, interact with the public, and carry out detailed work instruetions. Moderate is defined as limited but satisfactory.

We first consider whether this evaluation ignored three pieces of medical evidence. 1 First, Valentine points to the report of Dr. LeBray, one of the psychologists who reviewed Valentine’s medical record on behalf of the SSA. Dr. LeBray stated that Valentine needed “simple, paced (unrushed) tasks/routines without close public interaction.” Valentine claims that this limits him to “simple” work, whereas the RFC limits him to “medium exertion work.” However, Dr.

LeBray submitted a more in-depth “Mental Residual Functional Capacity Assessment,” which makes it clear that he did not intend to limit Valentine strictly to simple work, as that term is understood in the parlance of the SSA. The assessment describes Valentine as “[mjoderately [l]imited” in four of twenty categories of mental activity and “[n]ot [significantly [limited” in the other sixteen.

The four categories of moderate limitation were: “ability to carry out detailed instructions”; “ability to maintain attention and concentration for extended periods”; “ability to interact appropriately with the general public”; and “ability to set realistic goals or make plans independently.” The first three limitations apply to functional capacity; they appear almost verbatim in the RFC. Thus, we cannot conclude that the RFC ignores Dr. LeBray’s medical evaluation. 2 Valentine also claims the ALJ ignored or contradicted Dr. Storzbach’s neuropsychological assessment. Dr.

Storzbach summarized his assessment in several distinct sections. One section, titled “Impression,” states that the “[neuropsychological assessment suggested average baseline cognitive ability. Performance on many tests was within normal limits or better.... Results of some tests indicated variably impaired performance on measures of attention, working memory, and complex psychomotor function.” There is also a more detailed

summary of the test results, which notes that “the patient’s pattern of moderately impaired initial presentation performance on verbal memory tests suggested deficient attention.” With respect to “[djurability of memory for newly learned information during distraction,” “multiple measures derived from this test rang[ed] from normal to impaired.” These findings suggest some moderate neuropsychological impairments.

As such, they are consistent with the RFC, which included “moderate [ (meaning “limited but satisfactory”) ] restrictions of [Valentine’s] capacity to concentrate, interact appropriately with the public, and carry out detailed instructions.” Once again, the RFC actually incorporated the evidence that Valentine argues it ignored. Valentine asserts, however, that Dr. Storzbach also limited him to “highly routinized, overlearned tasks with low cognitive demand.” To be sure, the doctor noted that Valentine “is less likely to have difficulty with[such tasks].” But this notation appeared in a

section of Dr. Storzbach’s report entitled “Recommendations.” Nowhere in this

section does Dr. Storzbach indicate that Valentine is incapable of working except under the recommended conditions. Indeed, he pointed out that Valentine’s “mostly normal test performance with multiple cognitive strengths suggests that [he] is capable of at least partially compensating for his deficits.” Thus, we agree with the SSA that Dr. Storzbach’s observation about “highly routinized, overlearned tasks with low cognitive demand” is neither a diagnosis nor statement of Valentine’s functional capacity. It is rather a recommended way for Valentine to cope with his PTSD symptoms.

The ALJ therefore did not err by excluding it from the RFC. In sum, we cannot say that the ALJ ignored evidence of Valentine’s impairments when she fashioned his RFC. B Valentine also argues that the ALJ improperly rejected other evidence regarding the extent of his ailments: Dr. Van Male’s testimony, his own testimony, and his wife’s testimony. Our Social Security precedents have developed a highly articulated set of standards for reviewing an ALJ’s decision to reject different types of testimony.

We take each piece of testimony separately, as each invokes a different standard. 1 There are three types of medical opinions in social security cases: those from treating physicians, examining physicians, and non-examining physicians. Lester v. Chater, 81 F.3d 821, 830 (9th Cir.1995). Where a treating or examining physician’s opinion is contradicted by another doctor, the “[Commissioner] must determine credibility and resolve the conflict.” Thomas v. Barnhart, 278 F.3d 947, 956-57 (9th Cir.2002) (internal quotation marks omitted).

However, to reject the opinion of a treating physician “in favor of a conflicting opinion of an examining physician[,]” an ALJ still must “make[ ] findings setting forth specific, legitimate reasons for doing so that are based on substantial evidence in the record.” Id. at 957 (internal quotation marks omitted). Valentine objects to the ALJ’s rejection of the opinion of Dr. Van Male, who was Valentine’s treating psychologist. The ALJ gave more weight to Dr. Storzbach’s neuropsychological evaluation, as well as to the functional capacity evaluation of Dr. LeBray.

Indeed, as we explain above, supra at 690-92, the RFC clearly reflects the influence of Drs. Storzbach and LeBray, who both concluded that Valentine had various moderate limitations. Furnished with the contradictory opinion of an examining psychologist (Dr. Storzbach), the ALJ must have provided “specific and legitimate reasons that are supported by substantial evidence in the record,” for rejecting Dr. Van Male’s opinion. Lester, 81 F.3d at 830. We believe the ALJ met this standard. She identified a contradiction in Dr.

Van Male’s opinion, in that the doctor “repeatedly reported [Valentine] was unemployable while acknowledging he was continuing to work full-time.” The ALJ also noted evidence in the record, including Dr. Van Male’s “own treatment progress reports” that showed Valentine’s “improved functioning at work and encouraging comments he received from company officials.” This record evidence was consistent with the opinions of Drs. Storzbach and Le-Bray. Thus, the ALJ rejected Dr.

Van Male’s opinion for specific and legitimate reasons supported by substantial evidence. 2 Valentine makes a brief argument that the ALJ improperly rejected his own testimony about his pain and fatigue. He states that the ALJ’s “reasons for rejection are based largely on factual inaccuracies” and that his statements are consistent and corroborated by lay witness testimony. Under our case law, [without affirmative evidence showing that the claimant is malingering, the Commissioner’s reasons for rejecting the claimant’s testimony must be clear and convincing.

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. The ALJ must specifically identify what testimony is credible and what testimony undermines the claimant’s complaints. In this regard, questions of credibility and resolutions of conflicts in the testimony are functions solely of the Secretary. Morgan v. Comm’r of Soc. Sec. Admin., 169 F.3d 595, 599 (9th Cir.1999) (citations omitted).

As there was no evidence of malingering, the ALJ had to explain why she did not find Valentine’s subjective contentions about his limitations to be persuasive. The ALJ determined that Valentine “demonstrated better abilities than he acknowledged in his written statements and testimony” and that his “non-work activities ... are inconsistent with the degree of impairment he alleges.” Much as the ALJ did in Morgan, the ALJ here “pointed to specific evidence in the record,” id. at 599, that undermined Valentine’s claims that his PTSD was so severe he was unable to work.

While she recognized that “the treatment and employment records reveal that [Valentine] struggled in his traditional work” and that “[h]is job performance clearly suffered substantially,” the ALJ observed that the same records witnessed Valentine’s ability “to rally and improve his functioning.” In addition, the ALJ remarked on the fact that Valentine exercised and undertook several projects after he retired, including gardening and community activities.

The ALJ recognized that this evidence did not suggest Valentine could return to his old job at Cummins, but she thought it did suggest that Valentine’s later claims about the severity of his limitations were exaggerated. We conclude that, in light of Valentine’s conclusory argument, the ALJ provided clear and convincing reasons to reject his subjective complaint testimony. The ALJ identified evidence that “undermine[d] [Valentine’s] complaints,” id. at 599, and found such evidence credible. This evidence directly contradicted Valentine’s contentions about how debilitating his fatigue was.

Thus, we conclude that the ALJ’s resolution between conflicting evidence provided a clear and convincing reason to reject Valentine’s subjective testimony. 3 Valentine also argues that the ALJ did not sufficiently justify her rejection of the testimony of his wife, Tamara Valentine, and of Lane Anthony, his supervisor at Cummins during 2003 and 2004. When an ALJ discounts the testimony of lay witnesses, “he [or she] must give reasons that are germane to each witness.” Dodrill v. Shalala, 12 F.3d 915, 919 (9th Cir.1993). Mrs.

Valentine’s testimony of her husband’s fatigue was similar to Valentine’s own subjective complaints. Unsurprisingly, the ALJ rejected this evidence based, at least in part, on “the same reasons [she] discounted[Valentine’s] allegations.” In light of our conclusion that the ALJ provided clear and convincing reasons for rejecting Valentine’s own subjective complaints, and because Ms. Valentine’s testimony was similar to such complaints, it follows that the ALJ also gave germane reasons for rejecting her testimony. The ALJ also relied in part on two facts quite common to spouses: that Mrs.

Valentine, as Valentine’s wife, was an interested party, and that she never saw him at work. Such a broad rationale for rejection contradicts our insistence that, regardless of whether they are interested parties, “friends and family members in a position to observe a claimant’s symptoms and daily activities are competent to testify as to [his or] her condition.” Dodrill, 12 F.3d at 918-19. Although spouses usually do not observe each other at work, they do usually observe each other at home. Thus, insofar as the ALJ relied on characteristics common to all spouses, she ran afoul of our precedents.

This does not mean an ALJ must accept the testimony of a spouse who knows little about a claimant’s functional capacity. But the ALJ must explain such ignorance in the individual case. Similarly, evidence that a specific spouse exaggerated a claimant’s symptoms in order to get access to his disability benefits, as opposed to being an “interested party” in the abstract, might suffice to reject that spouse’s testimony. In any event, the error was harmless in this case. Nonetheless, we remind ALJs to tie the reasoning of their credibility determinations to the particular witnesses whose testimony they reject.

Finally, we address the ALJ’s rejection of the letter of Lane Anthony, Valentine’s supervisor, retracting the glowing employee reviews he wrote in Valentine’s last days as an employee at Cummins. Anthony claimed he lied for Valentine out of pity. But the ALJ found it more likely that Anthony’s professed sympathy for Valentine explained his desire to present him in the best light at the administrative hearing. After all, the ALJ reasoned, Anthony is no longer responsible to Cummins to provide truthful reports, as he was when he penned the performance reviews.

Of course, Anthony was also under a responsibility not to misrepresent facts to the ALJ. Nonetheless, as between two contradictory characterizations of Valentine’s work performance, the ALJ provided a reason germane to the witness for relying on the original performance reviews. C We conclude that, in fashioning the RFC, the ALJ neither improperly rejected nor ignored evidence. Therefore, the ALJ properly relied on the vocational expert’s responses to the hypothetical, which reflected the RFC. V Valentine’s final argument presents us with a novel issue.

While his case was pending before the ALJ, the VA rated Valentine 100 percent disabled. He argues that the ALJ ought to have accepted such rating as determinative here. We have held that “an ALJ must ordinarily give great weight to a VA determination of disability.” McCartey v. Mas sanari, 298 F.3d 1072, 1076 (9th Cir.2002) (reversing a denial of benefits because the ALJ “failed to consider the VA finding and did not mention it in his opinion”).

Nevertheless, “[bjecause the VA and SSA criteria for determining disability are not identical,” we have allowed an ALJ to “give less weight to a VA disability rating if he gives persuasive, specific, valid reasons for doing so that are supported by the record.” Id. Unlike the ALJ in McCartey, the ALJ here considered, but rejected, the VA’s disability finding. Thus, we must decide whether the ALJ met McCartey’s standard for doing so: “persuasive, specific, valid reasons ... supported by the record.” The ALJ discussed the VA’s disability rating in her evaluation of Dr.

Van Male’s opinion that Valentine was unemployable: The doctor’s advocacy efforts helped the claimant gain an initial increase in [ ] his disability percentage rating in 2004 and eventually gain a 100 percent rating....

While the VA unemployability rating resembles the Social Security disability standard in some respects, the non-critical decision made by the VA Decision Review Officer in May 2006 (granting a 100 percent disability rating and obviating the unemployability determination), was not an unemployability assessment and was not based on a comprehensive evaluation of the evidence available to the undersigned [ALJ], We discern two parts to this explanation.

The ALJ first suggests that the decision of the VA to grant 100 percent disability, as opposed to ruling on unemployability, is not relevant to a Social Security disability rating. This contradicts McCartey, which explicitly relied on “the marked similarity between [the disability programs of the VA and of the SSA].” 298 F.3d at 1076. Insofar as the ALJ distinguished the VA’s disability rating on the general ground that the VA and SSA disability inquiries are different, her analysis fell afoul of McCartey. The ALJ did, however, offer a second reason not to follow the VA.

She stated that the VA’s determination “was not based on a comprehensive evaluation of the evidence available to [her].” This explanation appeared in the same paragraph in which the ALJ justified her decision to discredit Dr. Van Male’s opinion. It is clear to us, therefore, that the ALJ was referring to the fact that the VA rested on an opinion that the ALJ rejected. Dr. Van Male’s opinion and Dr. Carter’s evaluation were important parts of the record before the VA.

We have concluded, supra at 692-93, that the ALJ provided “specific and legitimate reasons that are supported by substantial evidence in the record,” Lester, 81 F.3d at 830, for rejecting Dr. Van Male’s opinion. Valentine did not challenge the ALJ’s rejection of Dr. Carter’s opinion, which the parties agree relied on inaccurate information. Furthermore, the ALJ benefited from the opinions of the agency psychologists, evidence of Valentine’s work history and post-retirement activities, and the input of the vocational expert.

The ALJ was justified in rejecting the VA’s disability rating on the basis that she had evidence the VA did not, which undermined the evidence the VA did have. In other words, even though the VA and the SSA both determine whether someone is disabled, here the latter had evidence unavailable to the former. We thus conclude that, on this record, the acquisition of new evidence or a properly justified reevaluation of old evidence constitutes a “persuasive, specific, and valid reason[ ] ... supported by the record” under McCartey for according little weight to a VA disability rating.

VI For the foregoing reasons, we AFFIRM the judgment of the district court. . Anthony would later submit a letter to the SSA stating that he only gave Valentine positive performance reports because he pitied him and because he felt "there was no reason to kick a man on his way out.” . We also reject Valentine’s argument regarding the omission of his cervical and lumbar spine injuries and knee and shoulder injuries from the RFC. The cervical and lumbar spine injuries dated from Valentine’s service in Vietnam.

The record reveals no indication that they caused Valentine problems after March 30, 2004, the date that the alleged disability began. As for the knee and shoulder injuries, Valentine does not explain his contention that the ALJ did not account for these impairments. The ALJ concluded at Step 2 that they were severe, and the RFC includes several physical limitations. Valentine does not detail what other physical limitations follow from the evidence of his knee and should injuries, besides the limitations already listed in the RFC.

We reject any invitation to find that the ALJ failed to account for Valentine’s injuries in some unspecified way.