Coley Copeland, Jr. appeals the denial of his application for social security disability benefits. Finding that there was substantial evidence to support the Secretary's determination that Copeland had a residual functional capacity for some light and sedentary work and was therefore not disabled, we affirm.

* Copeland filed his application for disability benefits on March 29, 1985, claiming that he was permanently and totally disabled by the combined effects of a severe back injury, hypertension and nervous anxiety. He has not worked since January of 1980, when he left his position as a processor in a peanut packaging factory. Copeland last met the special earnings requirement for benefits eligibility on March 31, 1985.

The administrative law judge assigned to Copeland's case found that, as a result of a severe, work-related back injury sustained in December of 1979, the claimant could no longer perform his past relevant work as a peanut processor--a position which frequently required him to lift more than 100 pounds at a time. Based on the available medical evidence and the testimony of a vocational expert, however, the ALJ also found that Copeland's conditions were of insufficient severity to constitute a "presumptive" disability under the regulatory listings,1 and that the claimant in fact had a residual functional capacity for some unskilled light or sedentary work. Giving due weight to the evidence of pain-induced nonexertional limitations, and using the relevant grid rule as a "framework for decisionmaking,"2 the ALJ in turn recommended a finding of "not disabled." The Appeals Council later denied Copeland's petition for review of this recommendation, which thereby became a final decision of the Secretary of Health and Human Services. The district court affirmed, and this appeal followed.

The claimant challenges the ALJ's findings on three independent grounds, which we consider seriatim.

Copeland argues first that, in light of his treating physicians' consistent recommendations that he not lift heavy objects, the ALJ erred by finding him capable of performing the tasks required by light or sedentary work. By the terms of the Secretary's regulations, "[l]ight work involves lifting no more than 20 pounds at a time[,] with frequent lifting or carrying of objects weighing up to 10 pounds," 20 C.F.R. Sec. 404.1567(b), while "[s]edentary work involves lifting no more than 10 pounds at a time and occasionally lifting or carrying articles like docket files, ledgers, and small tools." Id. Sec. 404.1567(a). Here, the ALJ found that the claimant could lift and carry up to 20 pounds, and that he therefore had a residual functional capacity for the "full range of light work ... [,] reduced [only] by his need to change positions approximately every two hours and his need to avoid complex tasks." Record on Appeal at 34.

As indicated, Copelant's primary complaint is that this characterization of his residual "lifting capacity" is wholly inconsistent with the reports provided to the Secretary by his treating physicians. In our view, however, there was in the record "substantial evidence" to support the ALJ's finding that, throughout the period during which the claimant was still elegible for benefits, he could lift and carry up to 20 pounds. From 1980 to 1984, the claimant was treated by Dr. Thomas J. Arkins, a board-certified neurasurgeon. In February of 1980, Dr. Arkins performed a laminectomy to treat a herniated lumbar disk discovered in the claimant's back. Shortly thereafter, Dr. Arkins told the claimant that he could return to work, but also advised him not to lift heavy weights or engage in any other particularly strenuous activity. This of course lends credence to Copeland's claim that, at least initially, his back problems precluded even the minimal lifting required by "light work." Dr. Arkins' later reports suggest, however, that the claimant's condition improved considerably as a result of the corrective surgery. In May of 1980, for example, Dr. Arkins reported that Copeland had recovered sufficiently that "his only limitation would be a weight-lifting limitation of 10-20 pounds." Id. at 247.

Over the next twenty-four months, Dr. Arkins saw the claimant repeatedly. On several occasions, he advised that Copeland lift no more than 10 pounds. See id. as 232, 237 & 243. In a March 1983 report to the Suffolk, Virginia Department of Social Services, moreover, Dr. Arkins indicated that the claimant was no longer capable of lifting objects of any weight. Id. at 234. Dr. Arkins also consistently opined, however, that Copeland was "capable of working" - with certain "modif(ications in) the kind of work he was doing " before the December 1979 injury - and that the claimant was therefore "not totally disabled." Id. at 245. More significantly, another treating physician found Copelant capable of lifting up to 20 pounds as late as November 1985 - that is, several months after the claimant's elegibility for disability benefits expired. See Report of Dr. Lawrence Morales, Record on Appeal at 286.3 Indeed, the claimant himself reported in late 1985 that he could lift objects weighing up to 25 pounds. Record on Appeal at 134.

In the face of a record replete with conflicting medical opinions and recommendations, we owe considerable deference to the ALJ's findings of fact. See, e.q., Estep v. Richardson, 459 F.2d 1015, 1016-17 (4th Cir. 1972). It suffices to say that such is the case here; and, finding that it was supported by "substantial evidence," we therefore will not disturb the ALJ's finding that the claimant had a continuing ability to lift up to 20 pounds, hence to perform the tasks required by "light work."

Second, Copeland argues that the ALJ erred by "discountin" a state vocational rehabilitation agency's evaluation of his residual functional capacity. In June of 1983, the Virginia Department of Rehabilitative Services referred the claimant to the Woodrow Wilson Rehabilitation Center for an evaluation of his capacity for vacational "retraining." Woodrow Wilson consultants ultimately reported that the claimant could not perfor the tasks required by a number of the jobs in which he expressed interest, and recommended that he seek psychiatric treatment and academic training before attempting to return to the workplace. Copeland now claims that this report constituted affirmative evidence of total disability, and that the ALJ should have considered it dispositive.

For several reasons, we disagree. First off, the claimant mischaracterizes the report as concluding that he was completely disabled. As the Secretary points out, the Woodrow Wilson consultants found only that Copeland was a marginal candidate for retraining - and never suggested that he was incapable of securing gainful employment in an "unskilled" field. To the contrary, the report affirmatively suggested that the claimant had "no detectable motivation .... I felt he really wasn't reying(, and) I would say his motivation (was) microscopic." Record on Appeal at 219. Of course, this does not constitute positive evidence that, contrary to his claims, Copeland was not disabled; but it obviously undermines any support that the report might otherwise offer his case.

Perhaps most significantly, other evidence in the record stands in stark contrast to the claimant's generous interpretation of the Woodrow Wilson consultants' findings. At the original administrative proceeding, the ALJ heard Dr. George Jarrell, a consulting vocational expert. Dr. Jarrell opined that, even given the exertional limitations diagnosed by Dr. Arkins and the marginal prospects for successful retraining described in the Woodrow Wilson report, there were nevertheless a number of unskilled jobs which Copeland could perform. Dr. Jarrell suggested, for example, that the claimant could work as a production packager, locker room attendant or bridge tender. The ALJ was entitled to reply upon this considered opinion, and indeed to consider it consistent with the state agency's report. We simply cannot say, therefore, that the ALJ mischaracterized or failed to give adequate weight to the available evidence of Copeland's assertedly poor prospects for successful vacational rehabilitation.

Finally, the claimant argues that the ALJ erred by discounting his testimony that he suffered from debilitating pain. In particular, he characterizes as "sit and squirm" jurisprudence the following passage from the ALJ's final opinion and order:

The claimant has testified to the existence of pain. While pain can be disabling, ordinary subjective complaints of pain must be corroborated by some objective medical evidence of record. The record does show that the claimant has a history of muscle spasms and restriction of motion in the lumbar spine, which would cause him some discomfort. However, ther have been no indications of impairment in general nutrition, debilitation of physical well being, or other physical findings that normally accompany severe pain. In addition, the undersigned has had the opportunity to observe the claimant at the hearing where his attention was unimpaired .... (Moreover, psychological) testing indicated that the claimant had no significant impairment of his concentration. Therefore, his pain does not substantially reduce his capacity for performing light work.

Record on Appeal at 33 (emphasis supplied).

We agree with the claimant that the ALJ's reference to and apparent reliance on his observations at the administrative hearing was improper. "In addition to the obvious danger of unreliability, such (referenced) 'may encourage claimants to manufacture convincing observable manifestations of pain or, worse yet, discourage them from exercising the right to appear before an Administrative Law Judge for fear that they may not appear to the unexpert eye to be as bad as they feel.' " Jenkins v. Bowen, No. 86-3969, slip op. at 3 (4th Cir. May 18, 1987)(unpublished)(quoting Tyler v. Weinberger, 409 F. Supp. 776, 789 (E.D. Va. 1976)). See also Freeman v. Schweiker, 681 F.2d 727, 731 (2d Cir. 1982). We are also persuaded, however, that the error was harmless. Here, the ALJ also relied on independent, objective evidence that the claimant's pain, though perhaps sometimes severe, was not debilitating. Dr. Arkins various reports repeatedly stated, for example, that Copeland's spinal discomfort was not so serious as to constitute a total disability. One psychologist's report took special note, moreover, of the claimant's "(r)elative strength ... with tasks (requiring) concentration." Record on Appeal at 331. On the basis of this evidence alone, the ALJ was entitled partially to discredit the claimant's subjective pain testimony; and we therefore find no reversible error in his concededly improper, additional reference to Copeland's demeanor at the administrative hearing.

For the foregoing reasons, the judgment of the district court is affirmed.

AFFIRMED.